U.S. Environmental Protection Agency
Office of Administration and Resources Management
Office of Information Resources Management
Washington, D.C. 20460

          VOLUME 2:

      NCC-IBM  User's  Guide
      NCC-VAX  User's Guide
      LAN Technical  Guidlines

            NCC-IBM USER'S GUIDE
                February 1982
            Revised January 1988
                Prepared by=

The NCC-IBM User's Guide provides a useful source of
information about the center, its procedures, available
resources, and supported softuare.

The current version includes information published in User
Jlemos through January 1, 1988.

NCC-IBM USER'S GUIDE                                     JAN  1988


The NCC-IBM User's Guide was designed to be an introduction to
the NCC-IBM as well as a technical document.  Sections  1.0
through 3.0 address the more general information that a user
needs to know.  Sections M.O through 7.0 address the software
packages that are available and their more technical aspects.
The Appendixes contain information which is very dynamic in
nature and which may be too detailed to include in the  text of
the Guide, such as error messages or lists of commands.  The
Index is a comprehensive one which lists each subject in as many
forms as possible.


The purpose of the NCC-IBM User's Guide is to provide a useful
source of information about the center, its procedures, available
resources, operations, and supported software.  Much of the
information in the Guide about the IBM 3090-300, its utilities,
and its job control language is explained in greater detail in
various IBM manuals.   Extensive references to those manuals have
been included, and the user may obtain his own manuals  for
referencing (see Section 2.M.2.1).  Some utility programs have
been written at NCC and are not generally available at  other IBM
installations.  These are documented for the user in the NCC-IBM
User's Guide and may also appear in other NCC-produced  manuals.

The NCC-IBM User's Guide is--just as its name implies — a USER
document.  It is with this concept in mind that the NCC welcomes
any suggestions, comments, or contributions from its users.
Appendix B contains a Reader's Commentary form.


1.2.1    Numbering Systems

Section numbers follow the decimal system, e.g., Section 3.0 may
have subsections 3.1, 3.2, etc. and subsection 3.1 may  further
divide into Sections 3.1.1, 3.1.2, etc.  The standard in this
manual is to keep subdivisions to no more than four digits; e.g.,

Figure numbers follow the format n-n; e.g., Figure 2-7.  The
first number of a figure is the section number; the second number
is  the figure number within that section.

Page numbers are also denoted by section and page within that
section.  For example, the number 5-9 at the bottom of  a page
would indicate that the section number is 5 and the page number

NCC-IBM USER'S GUIDE                                     JAN  1988

within that section is 9.  When updates are issued, page numbers
may show an additional number; e.g., page 5-22 may be replaced in
an update with pages 5-22.1 and 5-22.2.

1.2.2    Jobstreams

The following conventions are used in jobstream examples:

     a.  All literals which are required are in upper case
         letters, such as DISP=, EXEC, etc.

     b.  User-supplied information is indicated in lower
         case letters, such as stepname, acct, or data-set-
         name .

     c.  Data card (or tape) input is indicated by the words
         which are indented and enclosed in parentheses,
         such as (control cards).

1.2.3    Submitting Comments

Users may contribute their suggestions on a Reader's Commentary
Form (sample shown in Appendix B).  The user should indicate
specific pages and paragraphs or submit copies of pages which he
has annotated.   Suggestions and corrections will be reviewed and
incorporated into the manual in the form of updates, as


A schedule has been established for the preparation of NCC-IBM
User's Guide updates.  Every April 1 and August 1 revisions will
be made available on-line for user retrieval.  The revision
cycles, identified as UPDATE1 and UPDATE2, respectively, will be
identified on revision pages with a bar (I) in the left margin
beside revised text.  A NEWS ALERT and a User Memo will announce
availability of updates.

The entire manual will be revised and reprinted each December 1
incorporating all revisions from the April and August revision
cycles.  The manual will  be reproduced on both sides of 8-1/2" x
11" three-hole paper suitable for insertion in a three-ring
binder, and will be mailed free of charge to users whose User
Communications System profiles reflect their requests for NCC-IBM
User's Guide distribution.  A sample User Profile Worksheet is
available in Appendix B.

Users may also print their own NCC-IBM User's Guide from the on-

NCC-IBM USER'S GUIDE                                      JAN  1988

line data set JUSD.USERS.GUIDE(MANUAL).  See Appendix  H  for
printing instructions.

NCC-IBM USER'S GUIDE                                     JAN  1988



The NCC is one of the largest, most modern, high-speed computer
centers in the nation. Its mission is to support the
Environmental Protection Agency in the areas of both scientific
and administrative applications. The NCC serves EPA users in  the
Washington, D.C.  Headquarters, Regional Offices, and laboratories
throughout the U.S. Services are also provided for other
government agencies and contractors.  The NCC consists of certain
physical facilities, hardware, and software which constitute  the
central computer system. A communications network allows
distribution of computer services to remote locations. The NCC
also provides user support services to its users.

The NCC is located within the Environmental Research Center,
Research Triangle Park (RTP), North Carolina. The RTF is situated
within the geographical triangle formed by North Carolina State
University, Duke University, and the University of North Carolina
at Chapel Hill. The NCC occupies approximately 46,000 square  feet
within the EPA Environmental Research Center.

The National Computer Center operates four principal data
processing facilities:  an IBM 3090-300, an IBM 3090-200, a
SPERRY 1100/82, and a minisystems operation consisting of a DEC
VAX 11/780, a DEC VAX 11/785, and a DEC VAX 8600.   The management
of the systems is organizationally structured under the National
Computer Center,  within the National Data Processing Division
(NDPD) of the Office of Administration and Resources Management
(OARM-RTP).  NDPD is responsible for managing the computer
hardware, software, telecommunications network, and support
services for the three major EPA systems.  The major NCC
computers share the EPA workload.  This document describes the
facility and services of the NCC-IBM system only.


NCC operations are conducted at two locations =  the National
Computer Center at RTP, North Carolina, where the central
computer system is located, and the Washington Information Center
(WIC)  at EPA Headquarters. The Washington Information Center
provides EPA Headquarters' users with remote access to most
required facilities and services.

NCC-IBM USER'S GUIDE                                     JAN  1988

2.2.1    NCC Hardware

The NCC facility at RTF has the hardware necessary to handle  a
wide variety of input and output media.  Punched cards, tapes,
printed output, microfiche, and graphics may be utilized via
over-the-counter service.

The collection of data processing devices that comprise a
computer installation, along with the arrangement of those
devices, is generally known as hardware configuration.  The NCC-
IBM configuration consists of one processing unit (CPU), with two
integrated central processing units, central storage, channel
subsystems, channel paths, controllers, and input/output devices.    Processor

The processing unit is the controlling center. It provides
hardware for the execution of programs, main storage addressing,
and initiation of communication between main storage and the
input/output devices. The processing unit of the NCC-IBM system
is an IBM 3090-300 and consists of three integrated central
processors plus a vector processor.  The 3090-200 has two
processors.  When vector is on, you lose one regular processor.    Central Storage

Central storage provides the processing unit with directly
addressable fast-access storage of data. Beth data and programs
must be loaded into main storage (from input devices) before they
can be executed. The NCC-IBM configuration has 128 million bytes
on the IBM 3090-300.  The IBM 3090-200 has 64 megabytes of
memory.    Input/Output

I/O operations are handled by the channel subsystem in the
processor unit.  The channel subsystem provides 48 channels with
an upgrade capability to 64.  All channels can be configured for
block-multiplexer operations and as many as four channels can be
configured for byte-multiplexer operation.  Any channel not
needed for byte-multiplexer operation can be configured for
block-multiplexer operation.  The IBM 3090-200 has 32 channels.

Failing channels can be removed from the operating configuration.
As many as eight control units can be physically attached to a
channel, and each channel can address as many as 256 I/O devices.

NCC-IBM USER'S GUIDE                                     JAN  1988    Tape Accelerator

The NCC-IBM employs two STC 4800 tape accelerator for improved
performance of magnetic tape operations.  The accelerator is
installed on the channel path between the EXDC and the tape
controller units.   Its purpose is to reduce the input/output
bottleneck caused when the control unit and the channel path  are
busy with an operation.  (Data transfer would take place at tape
device speed -- up to  1250 KB/sec.)  The accelerator has a 64 KB
buffer for storage of user data which allows concurrent operation
of two or more tape drives on the same channel path.  In this way
the maximum throughput of a data streaming channel path can be
increased from 1250 KB/sec. to over 2.8 MB/sec.    Controllers

Control units, or controllers, process the requests received  over
a channel path, generally directing data to form an input/output
device.  Most input/output devices may be accessed through one of
two channel paths from the EXDC.  Such dual channel path access
prevents delays and repetition of operation caused by a busy
channel path or control unit.

The NCC-IBM hardware configuration for controllers consists of
standard control units and recently installed STC 8880 and 8890
storage control units for STC 8380 dual density disk units which
were also recently installed.  The basic model 8880 provides  two
storage directors with each director performing as a separate
control unit.  The 8880 interprets and executes commands from the
channel path and controls interfaces between the channel and  the
disk units. The 8890, which is an upgrade to the 8880, contains a
high-speed cache memory which stages and prestages data from  the
disk units.  This cache technology is designed to reduce access
time in a sequential read operation by providing a "look ahead
capability" and subsequent data transfer from cache memory at
channel transfer speed.    Disk, Drum, and Tape Devices

Disk and tape data storage devices belong to a class of machines
able to write and read information using the medium of magnetic
spots in a film of ferrous oxide.

There are eight drum devices used in the NCC-IBM configuration.
These devices are used by the operating system for functions such
as paging.

Two types of disk units are used at the NCC-IBM.  The first type
is the STC 8650 (analogous to IBM 3350) dual density disk drives

                                                         JAN  1988
with a storage capacity of approximately 635 million bytes of
information.  These devices have 19,069 bytes per track, 30
tracks per cylinder, and 1110 logical cylinders per pack.  The
second type is the STC 8380 dual density disk drive which has
M7,M76 bytes per track, 15 tracks per cylinder, and 885 cylinders
per pack.   The total byte capacity for an 8380 is 2.5 billion.
The 8380's not only offer greater byte capacity, but they also
offer a 50 percent faster access speed and 2.5 times the transfer
rate of the NCC's other disks.

Figure 2-1 shows the specifications of the direct access storage
units and performance characteristics by unit and type of
Model Number
and Unit
STC 4000(6)
solid state
STC 8380
No. of Bytes
Per Unit
1 1 million
635 million
2 . 5 billion
Seek Time in
Min Mean Max
7 18 30
3 16 30
Transfer Rate
in Bytes
Per Second
1 . 5 million
1 1 98 thousand
3 . 0 million
            Figure 2-1.  Direct Access Storage Units

                                              JAN 1988
Several types of tape drives are employed in the NCC-IBM system
configuration. Figure 2-2 provides the characteristics of each
type presently in use. When multiple density figures are given,
the drive is capable of reading or writing at either rate.
1600/6250 bpi
800 bpi
200/556/800 bpi
. 6/ . 3
. 6
in . /sec
in millis
2 . 1
4 . 4
1 . 0
2 . 2
Rewind Time
Full Reel
in Seconds
                Figure 2-2.  Types of Tape Drives
Plotting and Graphic Devices
Incremental plotters are available at both the NCC and the
Washington Information Center for graphic output.  They consist
of Model 925 Solid State Memory Controllers and Model 1051
Digital Incremental Drum Plotters manufactured by California
Computer Products, Inc. (CalComp).  The controllers include read-
only tape drives for reading 9-track, 1600 bpi plot tapes.

A plot tape is produced by executing any of the various CalComp-
compatible graphics software packages (as described in Section
7.4) available at the NCC.  This plot tape is then plotted off-
line on either of the two CalComp plotters.

The model 1051 plotter uses a stepsize with a two-axis resolution
of 0.001 inch or 0.0028 inch along a diagonal. The 1051 is a
four-pen plotter using roll paper 33 inches wide.
Auxiliary Devices
The NCC installation includes a number of auxiliary devices
important to overall operations but not directly concerned with
data processing.

NCC-IBM USER'S GUIDE                                     JAN 1988

     a.  Uninterruptible Power Supply (UPS) - The NCC-IBM
         computer system has been provided with a power
         supply system that regulates and filters commercial
         power and provides for continuous operation during
         short-term power failures.   The power supply system
         consists of a battery plant, motor generators, and
         a redundant uninterruptible power supply.

     b.  Tape Degausser - A tape degausser is a device that
         passes magnetic tape through a strong magnetic
         field, thereby erasing any information recorded on
         it.  The NCC provides a Weircliffe Magnetic Tape
         Bulk Eraser Model 8, manufactured by Amos of
         Exeter, Ltd., for erasing tapes.  This device
         ensures an erasure level of not less than 80
         decibels; normal computer tape recording levels are
         between 50 and 60 decibels. This erasure level
         complies with the requirements of the Privacy Act
         of 1964 and meets National Security Agency
         standards.  The degausser will erase a 2400-foot
         reel of computer tape in about 15 seconds.  This
         operation removes all data including any labels and
         header information.

     c.  Tape Cleaner/Evaluator - NCC has two
         cleaner/evaluators for magnetic tapes maintained at
         the central site.  Tapes are cleaned and/or
         evaluated by request or as needed (based on an
         established set of standards).    Computer Print and Card Read/Punch System

The printers, the card punch, and the card reader available to
the NCC-IBM system are described in Figure 2-3 below.  All of
these units are located at the central facility in Research
Triangle Park.

JAH 1988
IBM 32 1 1
3216 prin-fc train
(2) T11 up/low
(2) PI 1 UCS upper
2,000 lines per
minute, single-
IBM 3800
1 1 .Oin 20,040 1pm
8. Sin 19,350 1pm
7. Oin 18,936 1pm
5. Sin 18,036 1pm
3. Sin 15,780 1pm
IBM 3505
1200 cpm
IBM 3525
300 cpm
                  Figure 2-3.  Printing Systems
The IBM 3800 Printing Subsystem is a very fast, high-volume,
high-quality, computer output printer.  See Section 4.3.2 for
full discussion on its use.
2.2.2    HIC Hardware    Remote Job Control Configuration

The NCC's Washington Information Center Production Services at
EPA Headquarters provides over-the-counter batch submission and
output retrieval services for local users.  The site has one Data
100 Model 78-104 Terminal Control Unit (TCU) used to coordinate
functions on a Data 100 high-speed line printer, a card reader, a
card punch, and a 1600 bpi tape drive.  The printer located at
the WIC is a drum printer, with upper and lower case character
sets.  As over-the-counter batch submission, the WIC accepts
cards, card image tapes, and plot tapes.   Cards, printed
listings, and graphics may be received as output.  A public
terminal room providing user work areas,  low-speed terminals, and
keypunch facilities is also available.

The WIC has installed an IBM 4381 Model 11 system consisting of a
central processor, two auto-load vacuum tape drives, and two
high-speed, fixed disk drive units.  The system provides multiple
application usage including ISO, VTAM, and JES2 .  The IBM 4381 is
also in constant communication with the IBM 3090-300 at RTP
through the use of cross domains.  The 4381 currently has 20

NCC-IBM USER'S GUIDE                                     JAN  1988

communication ports available.

The Washington Information Center has installed two Digital
Equipment Corporation (DEC) Model 11/70 minicomputer systems  each
consisting of a central processor, card reader, tape drive, and
three disk drives. This system provides high-speed communications
to both the Sperry and IBM systems as well as dial-up
communications to other systems. The PDF 11/70 currently has  52
low-speed (300/1200 baud) communication ports.

The address and telephone number of the Washington Information
Center are listed in Appendix A.    Plotting and Graphic Devices

The Washington Information Center maintains a CalComp 925
controller and CalComp 1051 drum plotter.

Refer to Section 7.4 for information on NCC-supported graphics
software.    Datagraphix 9835 Laser Printer

Two Datagraphix 9835 laser printers are available at the WIC  for
high-quality, channel-connect printing from the IBM 3090-300  at
speeds up to 20,000 lines per minute. See Section 4.3.3 of this
manual for full discussion on their use.


Telecommunications on the NCC-IBM system is provided for
interactive teletype terminals and remote batch, RJE terminals.
This access is provided into the NCC-IBM system through COMTEN
and IBM 3725 front-end processors.  These processors operate  in
Emulation Processing (EP) mode and Network Control Program (NCP)
mode .

2.3.1     Remote Batch Access

NCC-IBM supports RJE access for terminals running HASP multi-
leaving and non-multileaving protocols.   Communications are
provided by leased AT£T provided circuits for high utilization
sites, dial-in WATS 4800 bps service for general use,  and TYMNET
access for sites needing support for more than RJE access.

NCC-IBM USER'S GUIDE                                     JAN  1988

2.3.2    Interactive Access

The IBM computer supports a network of 3270-type terminals.   This
service is provided to all the Regional Offices, EPA
Headquarters, and various contractor sites and labs.  This
service is available to users on leased telephone circuits and
dial lines via cluster controllers or via protocol converters
that allow for ASCII TTY terminals to emulate full-screen 3270's.

Telecommunication services are also provided to the Washington
Information Center's IBM computer via mainframe-to-mainframe
links.   This allows for exchange of information between machines
and provides a link to the user for access through either IBM

TYMNET provides NCC with international and national access for
interactive users.  This access is gained in over MOO cities  in
the United States through locally dialed numbers.   Users in
Washington, DC, RTF, NC, Regional Office areas, Las Vegas, and
Cincinnati are provided with non-TYMNET facilities for access.
Washington users gain access through multiplexers, and some RTF
users are provided direct access to the system.  Instructions on
identifying and using telecommunications numbers are in Appendix
E .
2.3.3    Terminal Procurement

In procuring low-speed and remote job entry devices for
connecting to the IBM and/or Sperry computers, the following
rules and procedures should be followed:

     a.  The user should select the type of equipment needed
         or NDPD will assist the user if the usage
         requirements are known.

     b.  The selected equipment should be evaluated by NCC
         Telecommunications for compatibility with the EPA

     c.  The user completes the purchase request and sends
         it to NDPD for final approval.   NDPD can be
         contacted if the user needs assistance or
         clarification on this process.

     d.  Usually a vendor is selected on the basis of the
         lowest price.  Third parties and other peripheral
         vendors offer considerable discounts over the same
         or comparable equipment from the manufacturer.
         NDPD can assist the user in locating the best price

NCC-IBM USER'S GUIDE                                     JAN  1988

Three documents, available from the ADP Coordinators, assist  the
user with the integration of IBM 31XX and 32XX equipment
technology into EPA's data processing environment.  These
documents refer to IBM devices by number; however, all references
include IBM plug-compatible devices that are marketed by a number
of other vendors.   Products from four other vendors are
acceptable plug-compatible products.

     o  EPA IBM-Compatible Video Terminal Selection and
        Support Guidelines.

     o  EPA IBM-Compatible Synchronous Terminal Standards

     o  EPA Asynchronous Terminal Standards for IBM 3270
        Protocol Conversion.

IBM 31XX terminal technology provides EPA users with full-screen
video display terminal and printers compatible with the IBM
System Network Architecture (SNA).  EPA is using SNA to
standardize its data communication network and provide full-
screen display service to all EPA users.

EPA IBM-Compatible Video Terminal Selection and Support
Guidelines provides a selection criteria for video terminals  that
connect to the EPA SNA network.

EPA IBM-Compatible Synchronous Terminal Standards provides
individual device specifications for procuring new IBM or IBM
plug-compatible 31XX or 32XX terminals, printers, and

EPA Asynchronous Terminal Standards for IBM 3270 Protocol
Conversion provides information about which existing
ASCII/asynchronous video terminals can be supported through IBM
3270 protocol converters.
2.3.4    Telecommunication Service Request

NDPD is responsible for providing telecommunication support to
the Agency.  The increased use of telecommunications has changed
the need for information concerning changes to the network.  A
new user of an IBM 3270 type terminal cluster must be uniquely
identified to the network.  NDPD must have more information
earlier in order to meet the needs of the user on a schedule
which is acceptable to both.

NDPD is taking several steps to formalize the procedure for the
user to request service changes.  The Telecommunications Service
Request (TSR) is the single document for requesting most types of
service changes.  Services and information which must be

NCC-IBM USER'S GUIDE                                     JAN  1988

requested and provided via this form include the following:

     o  Full-screen terminal support.

     o  Graphics terminal support.

     o  ASCII (TTY terminal support).

     o  Remote job entry registration.

     o  Terminal relocations.

The TSR will be the only method for requesting these services.
Note that the request must be approved by the ADP Coordinator.
Copies of the form are available through the ADP Coordinator  or
by telephone request to the Network Support Group (see Appendix A
for telephone number).  Copies are available as on-line print on
the IBM 3090-300 in the data set named JCMT.TSR.FORMS(TSR).   A
copy of the Telecommunication Service Request form is also
included in Appendix B of this manual.

The TSR will not be required for users of a dial-up TTY type
terminal requesting User-ID and access telephone number.

2.3.5    BITNET Membership

NCC is a member of BITNET, an international telecommunication
network that links data centers at hundreds of universities and
research centers.   Users can communicate with associates at other
BITNET nodes by using standard TSO commands to send and receive
information.  There is no cost associated with the use of BITNET
except for the cost of the TSO session.  Commercial use of BITNET
is strictly prohibited.

Users may send data sets or messages to other BITNET users
through TSO.  Several examples follow.  Note that nodeid is the
destination's BITNET nodeid and that userid is the individual's
userid to whom the information is being sent.

     o  TRANSMIT nodeid.userid

        Sends a message to another user.  Allows the user to
        input lines of data in full-screen mode or in line
        mode.  Terminated by PF3 or by entering a null line
        in line mode.

     o  TRANSMIT nodeid.userid DA(data-set-name)

        Sends a data set to another user by coding the data set
                              2-1 1

NCC-IBM USER'S GUIDE                                     JAN  1988

     o  TRANSMIT nodeid.userid DA(data-set-name) MSG

        Sends a data set accompanied by a message.

To receive data set and/or messages through BITNET, the TSO
command RECEIVE is used.

The RECEIVE command determines the type of data.  It displays
messages and allows opportunity for a data set to be saved to

The following information facilitates the use of BITNET:

     1.   To help ensure that data sets sent to other non-MVS
         nodes will be readable, transmit only sequential data
         sets.   Partitioned data sets should be transmitted
         only to MVS sites.

     2.   The maximum file size that can be sent via BITNET is
         300,000 bytes.

     3.   It is the user's responsibility to find out the nodeid
         and userid of the intended recipients (usually through
         verbal communication).

     i*.   The most current list of BITNET nodes is in

     5.   NCCIBM1 is the nodeid of the NCC IBM 3090.

     6.   Received data must be promptly accepted by the addressee
         through the use of the RECEIVE command in order to avoid
         deletion of the un-received "spool file."

fiuestions should be directed to User Support (see Appendix A  for
telephone number.)

2 . U . 1    Business Hours

The scheduled hours of operation of the various NCC facilities
are presented in the following table.  Times are in Eastern
Standard Time or Eastern Daylight Time, depending on the season.
Services can be made available during nonscheduled hours at the
rate  of $4500 per wall-clock hour through prior coordination with
the Operations Manager.

                                                JAN 1988

NCC Computer
TSO Services
NCC-IBM User Support
I/O Control (RTF)

                   24 hours
                   24 hours
                   8 AM-7 PM
                   24 hours
                   24 hours
                   24 hours
                   24 hours

24 hours
24 hours
24 hours
24 hours
24 hours
7 AM-6 PM

until 8 PM
until 8 PM
until 8 PM
until 8 PM
until 8 PM
                 Figure 2-4.  Hours of Operation
Exceptions to this schedule are as follows =

     o  ADABAS Regions may be stopped for maintenance on Tuesday from
        4 AM until 6 AM.

     o  Batch processing of jobs in queue may continue on Thursday
        from 4 AM until 6 AM while interactive and data communi-
        cations services are unavailable.

     o  Occasionally it will be necessary to utilize most or all of
        Sunday for disk reorganization or new operating system testing
        Changes to the schedule will be i:sted in the data set

     o  On Saturday and Sunday, the biweekly full pack backups will
        process concurrently with production which may delay jobs
        that access data sets on DASD volumes being copied to tape.

Additional information may be found on-line in data set
User Information Services
  Availability of Manuals
Information about ordering manuals is available for individual
users who wish to obtain a particular document.  For additional
information about technical and reference materials related to
the NCC-IBM system, contact NCC User Support (see Appendix A for
telephone number).   Copies of NCC-IBM User's Guide and other
internal documentation are provided to all registered users on-
line .

Users may submit individual purchase requests for IBM
documentation from their regional offices to the appropriate IBM

NCC-IBM USER'S GUIDE                                     JAN  1988    User Memos

When printed information must be disseminated to the user
community, selective distribution is made via automated User
Memos.   Information is classified and distributed according to
user profile keys in an on-line user name and address data base.
Copies  of automated NCC-IBM User Memos may be retrieved on the
IBM system from the news data set JUSD.NEWS as members NUMEMxxx
where xxx is the number of the particular memo.  For reference
purposes, member MEMOINDX contains a list of memos by number  and
title.   Three times a year information in NCC memos relevant  to
the NCC-IBM User's Guide will be incorporated as updates to the
User's  Guide.  These memos are then deleted from JUSD.NEWS.

Information from all user memos previous to the date in the
preface is incorporated in this User's Guide.  If you would like
information on the subjects for subsequent memos, refer to
JUSD.NEWS.  Refer to Appendix H for instructions on printing  any
of these members.    System Status Information

Information about the current status of the NCC system may be
obtained from the status phone maintained by the NCC-IBM
operations staff. Telephone numbers are listed in Appendix A.    Signon Announcements

A broadcast facility is provided that sends messages to each
terminal user signing on to TSO.  The same messages are sent to
JES2 remote batch work stations as a header page on each job
printout. User Support is responsible for generating these
broadcast messages.                     "•'

Broadcast messages usually contain information of immediate
importance to the general user community.  Typically, these
announcements pertain to significant events such as scheduled
downtime of services, system difficulties, or impending changes
which may affect users.

Any important information which should be broadcast to the NCC
user community should be brought to the attention of User
Support.    On-Line Usage Aids

A group of partitioned data sets has been established to provide
comprehensive descriptions of the various software packages

NCC-IBM USER'S GUIDE                                     JAN  1988

available on the NCC-IBM system.  Each has a member called
CONTENTS which lists all of the other members of that particular
partitioned data set (PDS).  The names of these data sets conform
to the standard JUSD.package.DATA where package is the name of
the software package.   For example, JUSD.CALCOMP.DATA(CONTENTS)
references the member called CONTENTS in the partitioned data  set
that presents a description of the CalComp software and provides
aids to facilitate its use.  The names of the various data sets
are given in the discussions of the software packages in Section

Data set, JUSD.UTILITY.DATA is similar to these although it is
not restricted to the use of any particular software package.  It
contains a number of jobstreams set up to perform commonly-used
functions, such as copying or restoring data sets.

In addition to these specific documentation aids,  other on-line
information is made available when possible.  For example, the
data set JUSD.SYSTEM.MESSAGES contains members documenting many
of the frequently encountered messages from different sources
such as the Tape Management System (IMS),  JES2, TELECOM, or ASM2.

To print any member of these aids, see Appendix H.

2 . M . 3    Consultation and Assistance

The NCC provides extensive resources for consulting with and
assisting users who have problems related to their ADP usage.
Although many of these resources are spread throughout the NCC
organization, the primary focus of these services is NCC User
Support. The major areas of consulting services and user
assistance are discussed individually in this section, primarily
to inform the reader of the type and extent of services that are
2 . M . 3 . 1     Telecommunications Assistance

The NCC  provides and maintains hardware and software components
required to support the local NCC low-speed interactive and
national remote batch network.  Continuous monitoring of the
network  is provided, and when necessary, software modifications
and hardware changes are made.  Where possible, attempts are made
to  anticipate and prevent communications problems along the
network.  Users experiencing communications problems should
contact  Telecommunications (see Appendix A for telephone number)
for assistance.  They will work directly with users to test lines
and interface low- and high-speed terminals.

NCC-IBM USER'S GUIDE                                     JAN 1988    User Support Services

The NCC has a comprehensive user support service, providing the
user technical assistance, problem diagnosis, solution, and
tracking as well as consultation with respect to all NCC methods,
procedures, and software needed to access and effectively use the
NCC ADP resources.  NCC User Support is the primary point of
contact between the NCC user community and the NCC-IBM facility.
Personnel requiring assistance or information regarding the NCC-
IBM can contact User Support either by phone, walk-in, or mail.
The scope of activities and services supported include:

         o   User assistance on facilities and technical matters
         o   Problem analysis and resolution
         o   User problem tracking through the CPU system
         o   System enhancement requests
         o   ADP refunds and billing information
         o   Approved special projects and studies for users
         o   On-line dissemination of technical information
         o   NCC user interface activities

NCC User Support is located at EPA National Computer Center (see
Appendix A for address and telephone number).

All user problems and questions are recorded through the NCC
Centralized Problem Management (CPM) system.  A Problem
Management Record (PMR) is logged containing the User-ID, user
name, telephone number, problem description, problem resolution,
and additional information necessary to keep a history of the
user problem.  Users are encouraged to make note of the PMR
number in case future contact with User Support is necessary on
that particular problem.

This tracking system benefits both the users and the NCC.  The
user is assured of his problem being resolved and documented when
he is given a problem number.  Additionally, the user has a
problem number to reference when recontacting NCC.  Through an
acceleration mechanism within the system, unsolved user problems
will be resolved in the most timely manner.  The tracking system
benefits the NCC by providing a record of all user calls for
historical purposes.

Problem analysis is initiated when anyone brings a problem to the
attention of the NCC User Support staff.  They will then
determine the cause of the problem and provide a solution to the
user.  In a case where the problem is due to a malfunction of
vendor-supplied hardware or software, that vendor will be
notified and a correction requested.  When such a correction is
implemented, the user will be notified.

Enhancement requests are initiated and processed in a manner

NCC-IBM USER'S GUIDE                                     JAN  1988

similar to problem analysis described above.  The user will be
informed of the scheduled implementation date if the enhancement
is accepted.  Otherwise, the user will be notified of the reason
for refusal.

User Services receives a list each day of the jobs that were
cancelled the previous day and notifies the appropriate user
organization.  User Services also determines if a refund is
warranted and initiates one at that time.  Additionally, any  user
may request a refund for the costs associated with a job which
fails due to an error by the operator, a hardware failure, or an
error in the systems software or one of the supported
applications packages.   Refunds are not granted for such
situations as the job being run at a higher priority to assure
its turnaround in a certain time period.  Refunds may also be
requested for miscellaneous charges such as disk rental,
delivery, tape usage, etc.  Requests for refund should be
submitted to NCC-IBM User Support.  See Appendix B for a sample
form.  The user will also be required to send all requested
documentation supporting his request to NCC-IBM User Support.
(Refer to Appendix A for address.)

The data set 'JUSD.BILLING.DATA' contains billing information.
Refer to member CONTENTS for a summary of the information
available in this data set.

Upon receipt of a refund request, NCC User Support will assign a
unique refund number to the request for the purpose of tracking
and monitoring.  The request will be investigated and, if
approved, the final dollar amount will be entered into the
accounting system for automatic refund during the next billing
cycle.  Users will be notified if their refund requests have  been

Refund requests are initiated for the following reasons:

     o   Job cancellation by operator.

     o   System problems.

     o   Printed output that was lost, or destroyed, or has
         wrong alignment, or bad print.

     o   Tape problems such as wrong tape mounted, retention
         problems, density,  hardware/software problems, TMS
         problems, tapes released to wrong person, or
         destroyed/mangled tape.

     o   JES problems

NCC-IBM USER'S GUIDE                                     JAN 1988

Refund requests may be approved for =

     o   Erroneous miscellaneous charges.

     o   Job failure due to operator, harduare, software, or
         supported application software error.

     o   Loading or rebuilding lost/destroyed files that are
         backed up by NCC-IBM.

If the initial job that failed was rerun successfully with no
changes, documentation from both runs must be submitted to
justify a refund request.   Refunds will be approved only for the
job that failed.

Refund requests may not be approved if;

     o   User does not submit requested documentation
         (output listings, core dump, terminal output, and
         tape dump) with refund request form.

     o   Specific information (job name, job number, date,
         and a brief explanation of what happened) is not
         provided on refund request form,  (see Appendix B
         for sample form).

     o   Job violates standards and procedures in user memos
         and/or NCC-IBM User's Guide.

     o   Refund is for under $1.00.

     o   Refunds are not submitted within 30 days of actual
         occurrence (including Saturdays).

     o   Job is rerun by NCC-IBM after correcting or
         applying a workaround to the problem which caused
         original to fail.

     o   Job was run with a message level to suppress
         messages; i.e. not MSGLEVEL=(1, 1) .

     o   Job is not restartable and abends.   Refund will be
         made up to that point where  job is  deemed

     o   Job has an unreasonably high CPU time estimate and
         goes into a loop.  Even if the immediate cause of
         failure was operator, hardware, or  software error.

     o   Job fails to run because it  was dependent on
         another job which did not run successfully.

NCC-IBM USER'S GUIDE                                     JAN 1988

     o   Dependent steps within a job have interdependent
         steps with no condition code testing.

     o   Errors were caused by tapes created other than at

     o   Job is run to reconstruct a lost or destroyed file
         when user is responsible for backing up.

2.M.3.3    Production Control Services

Production Control is an extension of the NCC Operations
Department, and provides all NCC users with a central-site
representative to initiate, monitor, and control production-
oriented utilization of the IBM 3090-300 computer system.
Production Control jobs are those that are completely documented
and that are required by a user on a periodic, timely basis, such
as data base updates or reports which are required as a result of
a data base update.

A user who wants a job incorporated into Production Control
should make a formal request for Production Control services to
the NCC Operations Technical Manager.  Thi request will be
evaluated in terms of man-hours, resources, complexity of the job
task, etc., and, if the request is approved, the user will be
contacted to provide specific details associated with the job

One of the first and most important processes in obtaining
Production Control services is the development of a runbook for
the job.  The runbook is the official guide from which Production
Control initiates, monitors, and verifies the job.  It is
produced by Production Control and the user, and it must contain
complete step-by-step job requirement and processing procedures.
It should also be explicit enough so that deviations from it are
seldom necessary.   In addition, the runbook will be used to
verify the accuracy and completeness of the job requirement and
processing procedures by executing a trial run of the job.  Upon
successful completion, the job will be accepted for Production
Control services.   Thereafter, any changes to the production
processing or requirements of the job must be provided to
Production Control by the user.

Production Control will initiate a job in accordance with the
established schedule or upon request from the user.  Production
Control personnel will not rearrange card decks or change any JCL
unless it is agreed upon with the user or it is stated in the
runbook.  In all cases, the user will assume the responsibility
should a change cause damage to a data set, file, or otherwise
create processing delays or problems with the job.

NCC-IBM USER'S GUIDE                                     JAN  1988

Production Control personnel will not create/generate card decks,
runstreams, or data files.  Nor will they be responsible for
punching cards except as stated above.  If a critical error
should occur during the processing of a Production Control job,
Operations will contact the Production Control representative
and/or the user.  The user may then correct the error(s) and  if
assistance is necessary, User Services Support should be

For more information and request procedures for Production
Control services, contact:

     Mr.  Tommie Rogers
     Environmental Protection Agency
     National Computer Center (MD-34)
     Research Triangle Park, NC  27711
     Phone:  (919) 541-2377 or CFTS) 629-2377    Education Services

Providing ways for users to learn about the various hardware
configurations and software services provided by the data center
is the mission of the NCC Training Office.   Two types of training
are provided:  (1) traditional classroom instruction, and (2)
media-based instruction.  At the NCC, available media-based
instruction consists primarily of Computer-Based Training (CBT)
and videocassette courses.  For more information about any of the
training services described below, call User Support (see
Appendix A for telephone number) and ask for the Training Office.

Classroom courses are scheduled on a quarterly basis at the NCC
in Research Triangle Park, NC, and at the Washington Information
Center at EPA Headquarters in Washington, DC.  The schedule is
published in a User Memo and is also distributed in leaflet form.
In addition to the published schedule, NCC training courses can
be presented upon request at any EPA sitr. .   Descriptions of
courses in the NCC training curriculum are provided on-line in
the data set JTNG.SYLLABUS and are also listed in Appendix K of
this manual.  To discuss scheduling a course at your location,
call User Support and ask for the Training Office (see Appendix A
for telephone number).

Advanced Systems, Inc.  (ASI) is presently under contract to
supply the NCC user community with videotape training.   The ASI
curriculum includes a variety of topics, including data
processing, management, and communication skills.  ASI catalogs
are provided to Regional ADP Coordinators,  or users may call User
Support and ask for more information from the Training Office
(see Appendix A for the telephone number).

NCC-IBM USER'S GUIDE                                     JAN  1988

Computer-based training on the NCC-IBM is provided via the
PHOENIX system.  PHOENIX includes a CBT course delivery system
and a coursewriting capability with the EASE CBT authoring
language.   PHOENIX courses are available to any user who has
access to the NCC-IBM system via a full-screen 3270-compatible
terminal.   Call User Support (see Appendix A for telephone
number) and ask for the Training Office for a PHOENIX student ID
and to register for a course.  A listing of all available CBT
courses is provided in Appendix K and in the data set
2 . M . M    Data Processing Support Services (DPSS )

2 . M . M . 1     Tape Library

DPSS  is  responsible for all computer tapes stored at the NCC.
This  responsibility includes inventory control, utilization and
ownership reporting, replacement of defective tapes, purchase of
new tapes, and initialization of volumes at regular intervals.
The tape librarian uses the on-line Tape Management System (IMS)
(see  Section 3.3.M for more details) in the performance of these

2.M.4.2     Disk Library

DPSS  is  responsible for providing inventory control, ownership
reporting, inspection and formatting of disk packs, and
coordination of all requests from users for new packs.   NCC Data
Management manages the contents, usage, and backup and recovery
of  all  disks.   The following sections describe specific functions
under these responsibilities.

2.M.M.3     I/O Control

DPSS  is  responsible for receiving batch jobs (card input) from
users.   The DPSS technician logs the jobs and reads them into the
system  via a card reader.

DPSS  is  also responsible for retrieving, separating, and
distributing output (cards, tapes, printed listings, and plots).

DPSS  is  also responsible for other I/O activities =

     a.   Receive, verify authorization of and coordinate
         requests for priority 5 processing. See Section

     b.   Prepare, log, and ship output.

NCC-IBM USER'S GUIDE                                     JAN  1988

     c.  Answer the computer room telephones.

     d.  Coordinate routine and special duties performed by
         the contracted courier service.

     e.  Receive foreign tapes, update the on-line system,
         assign slot numbers, inform users, and release and
         return tapes to users.

     f.  Distribute monthly tape and disk lists to users and
         ADP Coordinators.

     g.  Control visitor access to the computer room.

     h.  Distribute User Memos.

     i.  Process payroll output.

     j.  Coordinate billing for various systems.

2.M.M.4    Micrographics Services

Data Processing Support Services at the National Computer Center
and the Washington Information Center at EPA Headquarters are
responsible for receiving and monitoring requests from users  for
micrographics services.  Host requests for these services are
generated automatically by the user as the tape recording COM
data is created.  However, verbal requests are accepted.

The process is handled jointly by DPSS and Production Services.
Data sets recorded on tapes to be reproduced on microfiche must
begin with "COM.".

DPSS pulls these tapes from the library, updates the Tape
Management System to reflect the status of the tape as going
offsite for microfiche processing, files the tape to the
micrographics contractor.

To avoid confusion, local users who wish to send tapes offsite
for microfiche processing should either submit a COM request  form
or contact DPSS and provide the appropriate User-ID information.

Three times a day (at 0700, 1200, and 1500) Bulk Data Transfer
scans the system for the data set COM and User-ID's.  Only
specified User-ID's will print at NCC, the remainder will be
transmitted to the Washington Information Center.   The data sets
are then copied to tape and forwarded out to the micrographics

NCC-IBM USER'S GUIDE                                     JAN  1983

Current micrographic rates are listed in Section

A copy of the current NCC-IBM Micrographics Users Manual is in
Appendix F.    Keypunch Services

DPSS is responsible for receiving requests from the users for
keypunch services.  DPSS logs the request, pulls any requested
tapes, and forwards the request with specifications to the
keypunch subcontractor.  Upon completion of the job and its
return to the NCC, DPSS informs the user that the job has been
returned.  The user may retrieve the output or have it forwarded
to him.

The form used to request keypunch services is in Appendix B or
may be requested from DPSS.  fiuestions should be directed to DPSS
(see Appendix A for telephone number).

Current keypunch charge rates for specified turnaround requests
are listed in Section following.    Courier Service

DPSS is responsible for coordinating the courier service in the
Research Triangle Park area.  A commercial vendor has been
contracted to perform this service.  Twice each day, at 0700
hours and 1300 hours, DPSS dispatches output via courier to
offices located in the RTF area. Deliveries of input to DPSS from
users are made each weekday at  1100 hours and at 1600 hours.  RTF
area users may contact DPSS (see Appendix A for telephone number)
to request/coordinate this service.

Production Services at the Washington Information Center is
responsible for coordinating the courier service in the
Washington, D.C. area.  A commercial vendor has been contracted
to perform this service.  Twice each day, at 0630 and 1230 hours,
the WIC dispatches output materials via the courier to EPA and
contractor facilities throughout the Washington area.  In keeping
with the Data Center philosophy of charging users for service
performed, user accounts are charged for the expense of the
service.  The cost of providing the service is prorated to user
accounts based on the number of stops, the number of
organizations at each stop, and the number of accounts within
each organization.  Washington area users may contact Production
Services at the WIC (see Appendix A for telephone number) to make
arrangements for the addition or deletion of a stop.


2.4.5    Cost of Services
                                              JAN 1988
The cost of services at the National Computer Center is a
nonprofit chargeback system.  It was designed with a goal of
being fair, simple, predictable, and repeatable.
2 .4.5. 1
NCC Chargeback Algorithm
Dollars charged to users are determined by the resources used and
the level of service requested.  Resources are divided into
computer job components and computer-related resource components.
Computer job component costs are determined by the application of
the NCC chargeback algorithm.

NCC-IBM USER'S GUIDE                                     JAN  1988

The algorithm employed for batch jobs is as follows:

        Batch Job Cost = (CPU* + 1/0$) * PRF + U/R$


        CPU* = CPU Rate x CPU hours

CPU rates were established by machine: IBM 3090-300
                                            $900/CPU hour

                                       WIC 4381
                                            $93/CPU hour

                                       LMF sites
                                            $93/CPU hour

        I/O* is composed of one or more of the following:

        Tape Excp's = $0.45 per 1000
        Disk Excp's = $0.45 per 1000
        Terminal I/O = $0.45 per 1000
        PRF = priority factor as listed b^low:

                     PRTY       PRF

                       5        6.0
                       4        2.0
                       3        1.5
                       2        1.0
                       1         .5

        U/R = unit record as listed below:

              Local print lines = $1.50 per 1000 (includes WIC)
              Cards read = no charge
              Cards punched = no charge

Interactive (ISO and CICS) job cost algorithm is as follows:

NCC-IBM USER'S GUIDE                                     JAN  1988

       Interactive Job Cost = (CPUS + 1/0$)  * PRF + Connect*

where =

       CPU* = (TCB + SRB) hours * CPU hourly rate

       where:  TCB = Task Control Block
               SRB = System Resource Block

       Tape Excp's = $0.45 per 1000
       Disk Excp's = $0.45 per 1000
       Terminal I/O = $0.45 per 1000
       PRF = 1.0 for CICS and 1.0 for ISO
       Connect* = connect time at $9.00 per hour

Interagency/Intergovernmental Agreement (IAG) users are charged
an additional 10 percent of job component costs to recover
government labor and facility costs, and several other items
which are not required to be included in the computation of the
chargeback figure at this time.


2. M.S. 2    Charges for Computer-Related Services

Charges for computer-related services are listed below.
            JAN 1988

      Public On-Line Disk Storage
      Private On-Line Disk Storage
$9 . 00/Connect Hour
$.015/Track per Week
$1700/Pack per Month

      NCC-IBM Tape Rental/Storage
      Tape Mounts
      Work Tapes (creation and
          (expiration are same day)
$10.00/Tape per Month
No Charge

No Charge

      Days 1-5
      Over 5 Days

$2 . 00/Day

$5.00/Tape per Month

      Lines printed locally (RTF £ WIC)
$1 .50/1000 lines




      Cards read locally
      Cards read at user RJE
      Cards punched locally
      Cards punched at user RJE
                                    JAN  1988

                         FY87 CHARGES
                        No Charge
                        No Charge
                        No Charge
                        No Charge
MICROFICHE PRODUCTION (Subject to increase on 4/1/88.)
$3.19 per fiche + 10% Service Fee
$2.61 per fiche + 10% Service Fee
$2.15 per fiche + 10% Service Fee
*1.83 per fiche + 10% Service Fee
*1.65 per fiche + 10% Service Fee
$1.51 per fiche + 10% Service Fee
$0.165 per fiche + 10% Service Fee
First 50
Next 100
Next 200
Next 300
Next 600
All Others
      Special Delivery
      Data Set
        Dearchiving Assistance
           Cost + 10% Service Fee

                  $13.00/Data Set
(Suject to increase on 7/1/88.)
           Cost + 10% Service Fee
          12 hours
          24 hours
           3 days
           7 days
          14 days
           Cost per Thousand
        (does not include 10% handling)




             $50/Plotter Hour

          $260/Stop per Month

             $50/Staff Hour

          $1800/Site per Month

NCC-IBM USER'S GUIDE                                     JAN  1988



3.1.1    Registration for Accounting and Security

Registration procedures for setting up new accounts or adding
users to an existing account cannot be completed by the average
user; however, procedures are included for informational

To establish a new computer account, the form EPA ADP IBM, LMF,
SPERRY, and VAX Account and User Registration is required  (see
Appendix B for a sample).  The required signature of the EPA ADP
Coordinator denotes his approval to establish the new account
within his organization and that the correct Financial Management
System (FMS) accounting code has been supplied.  Agency ADP
timeshare  funds are allocated and monitored at the Program
Element (PE) level for each allowance holder.  Therefore, each
ADP account must be associated with a PE and an allowance holder.
The FMS accounting codes provide this required linkage to the
Agency's fiscal structure.  fiuestions regarding FMS codes should
be directed to EPA Senior Budget Officers or EPA Regional
Division Management Directors.

For Interagency/Intergovernmental Agreement accounts the EPA IAG
Coordinator's signature is required.  The user should contact:

       Linda Garrison
       EPA PM-211M, WSM
       401 M Street, SW
       Washington, DC  20M60

EPA ADP IBM, LMF,  SPERRY, and VAX Account and User Registration
form is supplied to all EPA ADP Coordinators through the TSSMS
Office and includes space for the user to be assigned to the new
account with special features.  When establishing a new account,
the signature of the ADP Coordinator is required.  This form is
also used  to add users to an existing computer account.  In
addition,  users may be deleted or changed and special features
can be provided to each user with this form.  When updating users
on an existing account, either the ADP Coordinator or the Account
Manager is required to sign.

EPA ADP Services Account Update form is for updating budgetary
information, deleting accounts, or changing the EPA Account
Manager.   The ADP Coordinator's signature is also required.

Both forms are provided to EPA Account (Project) Managers and ADP
Coordinators through the TSSMS Office.  Samples are in Appendix B
of this manual.

NCC-IBM USER'S GUIDE                                     JAN  1988

Temporary authorization can be obtained through the appropriate
EPA ADP Coordinator and/or EPA Account (Project) Manager by
contacting the TSSMS Office (see Appendix A for telephone number)
and requesting the valid information for the account.  After
receiving temporary authorization, the form N258 must be
completed and forwarded to the TSSMS Office within two weeks  for
the transaction to remain active.
3.1.2   NTIS Accounts

Non-Federal or non-Governmental agencies who need access to EPA
data or computing resources may do so through the National
Technical Information Service (NTIS).  NTIS sets up an account
through which billing for EPA services and computing resources is
accomplished.  The address for more information is as follows;

     U.S. Department of Commerce
     National Technical Information Service
     5285 Port Royal Road
     Springfield, VA  22161
     Telephone:  (FTS) 737-4807
                 (703) 487-4807

Prior to establishing an account, NTIS requires that an on-line
license/agreement form be completed by the requestor and then
received at NTIS.
3.1.3    User Identification Codes

Each account number may have multiple users validated for its
use.  A three-character User-ID is assigned to each of the
individual users.  This User-ID is in the format iii, where the
characters iii are unique to the user and assigned by the TSSMS
Office (these are often referred to as the user's "initials").
The NCC-IBM computer system requires users to enter a valid User-
ID when signing onto a terminal or when submitting a batch job.

When a user terminates employment, his User-ID will be deleted
from the system and after 45 days all data sets belonging to it
will also be deleted unless they are renamed.  To lessen the
impact of this restriction, group data sets should be used when
data files are being created for a large community of users .

Any users who sign onto the NCC-IBM system must be identified on
N258 for each account code (project) where applicable.  If any
person whose name is entered as a user under a particular account
code already has a User-ID under one or more other NCC-IBM
accounts, that fact should be noted in the block labeled "User

NCC-IBM USER'S GUIDE                                     JAN  1988

The name associated with a User-ID in the TSSMS file must be  the
valid/current user on the system.  The name assigned should be
that of the individual actually using the data sets or submitting
jobs, not a supervisor.  This person will be solely responsible
for that User-ID.  Any contact made by NCC concerning the use of
any User-ID will be made only to the person responsible for that

Use of a User-ID by two or more persons is discouraged except in
extenuating circumstances as determined by TSSMS.

All non-EPA personnel, such as contractors, must also have
individual User-ID's associated with each person's name.  The EPA
Project Officer's name or account manager's name should not be
used in association with these individuals' User-ID's.  The
ultimate responsibility for these non-EPA personnel User-ID's
remain with the EPA Account Manager (an EPA employee) associated
with the account under which the ID is registered on the Form

The names of all contractor personnel who will have access to the
NCC system must also be entered on the form with name of company
in user address and correct status marked.

The bimonthly Account Information Report, containing the current
N258 and N251 data, is sent to every Account Manager and ADP
Coordinator as a current record of the account and user
information (see Section 3.1.1).    Invalid User-ID

When it is discovered that the person listed as being responsible
for a User-ID cannot be located or no longer accepts
responsibility for the User-ID, the following action is taken:

     a.  The responsible EPA Account Manager is provided
         with notification of the above and is given 30 days
         to respond.

     b.  For security reasons, the User-ID is denied access
         to the system until the action to be taken
         concerning the User-ID is accomplished.   Should a
         user call who has been denied access to the system,
         TSSMS will, at the direction of the EPA Account
         Manager or EPA ADP Coordinator, establish temporary
         access which will be valid through the 30-day
         period specified in the notification. Unless
         otherwise notified, after 30 days, TSSMS will
         delete the User-ID.

NCC-IBM USER'S GUIDE                                     JAN  1988

     c.  The resources owned by that deleted User-ID are
         identified in writing to the responsible EPA
         Account Manager (see Section and those
         resources will be deleted from the system after 30
         days from the date of such notification if the EPA
         Account Manager does not respond or indicate that
         they should not be deleted.    Invalid Account

When it is discovered that the person responsible for an invalid
User-ID is an EPA Account Manager, or if it is brought to the
attention of TSSMS that an EPA Account Manager cannot be located,
the following action will be taken:

The responsible EPA ADP Coordinator will be provided with
notification that a new EPA Account Manager must be registered
with TSSMS within 30 days for the accounts for which the previous
EPA Account Manager was responsible.    Deleted User-ID or Deleted Account

whenever a User-ID or an account is deleted, the following
actions are taken:

     a.  All deleted User-ID's are removed from all
         documentation and mailing lists.

     b.  The responsible EPA ADP Coordinator or EPA Account
         Manager is notified of any tapes, private disk
         packs, and on-line data sets belonging to the
         deleted User-ID or account.  This written
         notification will allow him 30 days to dispose of
         these resources by reassigning tnem to a valid
         User-ID or account or by releasing them for reuse
         by the user community.

Failure to take action within the allotted 30 days in disposing
of these resources associated with a deleted account or User-ID
results in the following:

     a.  All tapes belonging to a deleted account and
         associated User-ID's are sent to the responsible
         EPA ADP Coordinator; a charge of *25 per tape is
         levied against an active account for which that
         Coordinator is responsible.  The EPA ADP
         Coordinator is notified by memo concerning the
         active account to which this charge is made.

NCC-IBM USER'S GUIDE                                     JAN  1988

     b.  All tapes belonging to any deleted User-ID are sent
         to the responsible EPA Account Manager; a charge of
         $25 per tape is levied against the account under
         which the User-ID created the tape.

     c.  Private disk packs associated with a deleted User-
         ID or account are copied to tape(s) and the disk
         packs released for further use.

     d.  On-line disk data sets associated with a deleted
         User-ID or account are automatically deleted, if
         any exist, 30 days after the EPA ADP Coordinator or
         EPA Account Manager has been notified in writing.

3.1.3    ADP Utilization Identifier

The ADP Utilization Identifier identifies a specific ADP system
or activity and associates computer usage statistics with that
activity.   A valid ADP Utilization Identifier is required for all
computer runs processed at the NCC.  The structure of the ADP
Utilization Identifier is detailed in Section M.1.1 with the
description of the JOB statement which requires it.  As indicated
there, the ADP Utilization Identifier contains a four-character
acronym which identifies a specific ADP system or activity.
These acronyms have been assigned to existing ADP systems and
activities.   Each user must obtain the appropriate acronym for
his use from his ADP Coordinator or Project Manager.  Buestions
pertaining to the ADP Utilization Identifier should be directed
to the Facility Impact Monitoring and Analysis System (FIMAS)
Office (see Appendix A for address and telephone number).


A basic premise of any ADP security effort is the provision for
strict individual user accountability and control in terms of
both machine and data access.   The security effort at the NCC
addresses  both user accountability and access controls in order
to provide the necessary protection.  All access to computer
services is controlled by restricting access to users who can
successfully identify themselves to the system.  Although the NCC
provides reasonable tools for the user to protect his data and
budget, it is up to the user to closely guard his access
authorization and avail himself of the data protection facilities

The Data Center also has the responsibility to provide the user
with the tools necessary to protect any proprietary, Privacy Act
(PAD, Business Confidential data that has been declared
sensitive  either by the user or by law.

NCC-IBM USER'S GUIDE                                     JAN  1988

It is difficult for the Data Center to say what is and what is
not sensitive; however, the Privacy Act of 1974 does spell out
that certain personnel data should be treated as sensitive, and
when dealing with it, the user has a responsibility to protect
that data.  According to the act (Public Law 93-597) PAI data
means any item, collection, or grouping of information about  an
individual that is maintained by an agency (of the Government)
which associates his/her name or an identifying number (i.e.,
social security number) to such information as the following:

     o  His/Her Education
     o  Financial Transaction
     o  Medical History
     o  Criminal or Employment History

In addition to the above categories, the following are also
considered PAI=

     o  Age (Date of Birth)
     o  Sex
     o  Handicap Code
     o  Veteran's Preference Code
     o  Employment History
     o  Service Computation Date
     o  Monitoring Group Designation
     o  Ethnic Background Indicator

These are examples of the type of information that the user must
protect from access by unauthorized personnel, but they are not
the only types of data that require protection.  It cannot be
stated too often that the user has the responsibility to
determine if his data is sensitive and proceed accordingly.

3.2.1    Obtaining and Changing Passwords

Passwords are established at the time the User-ID is established
and are given to the new user at that time.  When the new user
first tries to access the system, he must change his password
immediately by entering old password/new password when queried by
the system.  Unlike the User-ID and account code, a password may
be changed at any time.  Care should be exercised to avoid
revealing a password to unauthorized users; whenever it is
suspected that the secrecy of a password is compromised,  it
should be changed.  Only 10 consecutive logon attempts with an
invalid password are allowed before the User-ID is revoked by
RACF.  These attempts are not limited to any time period.  Each
successful logon reinitiates the logon attempt counter.  The
specific characteristics of the password are as follows:

NCC-IBM USER'S GUIDE                                     JAN  1988

     a.  Contains up to eight alphanumeric or national  ($,
NCC-IBM USER'S GUIDE                                     JAN 1988

3.2.3    Data Bases

NCC provides various data base products external to RACF, but
with interfaces to RACF, through which the user can store,
retrieve, and protect data.  Many of these products employ
internal security mechanisms through which data can additionally
be protected.

3 . 2 . M    Physical Security

Access to the computer room is limited to authorized personnel
only.  Entrance is controlled by a badge reader system.  Visitors
are allowed only when properly authorized by EPA management.
They must be escorted at all times by specially designated
personnel and must display the visitor's badge.

The computer room is environmentally controlled for both
temperature and humidity.  Doors must be closed at all times.
Food and drink are not permitted in the computer area.  A no
smoking rule is also in effect.

The computer room has special sensors to detect the presence of
smoke, fire, and water.  A detection of a hazardous situation
results in an alarm being sounded, and the alarm is pinpointed on
a central annunciation board.  In the case of a fire, a Halon
system is automatically dumped, unless manually aborted, to
contain the blaze.
3.2.5    Data Security

Users should note that a deletion of data from DASD does not
erase the data from the physical disk volume.  Similarly, when a
tape volume is eligible for scratch, routine erasing of the tape
(degaussing) does not take place.  It is possible in some cases,
although not routine or a common occurrence, for data to be
obtained from deleted disk files or scratched tapes either
intentionally or accidentally.  Users of sensitive data who are
concerned about the possibility of their data being disclosed
through such an occurrence should contact User Support (see
Appendix A for telephone number) who can assist them in further
protection of their data.
3.2.6    Disaster Recovery Plans

Disaster recovery procedures at the NCC include the backup of
critical data.  The NCC maintains an offsite vault which contains
backup material including tapes.   Data on user disk packs is
routinely backed up to tape and moved to this vault.

NCC-IBM USER'S GUIDE                                     JAN  1988

Documentation covering the specific concerns of disaster recovery
planning is available from the National Data Processing Division.

Users interested in preparing application-specific disaster
recovery plans should contact User Support who uill put them  in
touch with NCC disaster recovery personnel.  These personnel  can
assist the user and ensure that user plans are consistent with
NCC disaster recovery plans.


3.3.1    Data Set Naming Conventions

The data set name is assigned to a particular data set and serves
to distinguish one data set on a given volume, either a disk  pack
or a reel of tape, from all others on that volume.  The following
sections discuss standards set for the choice of data set names,
both those imposed by the Operating System and also by NCC
conventions.    Operating System (OS) Standards

The Operating System requires all data set names to be strings of
up to 44 characters. Standard names must also conform to the
restrictions outlined below.  The following discussion deals only
with standard data set names ••

     a.  The only permissible characters in a data set name
         are alphabetic (A through Z), numeric (0 through
         9), national (<3,  #, and $) characters and periods
         ( . ) .

     b.  The entire name must be divided by periods into
         segments, called  qualifiers, of up to eight
         characters.  For  example, given the data set name
         JUSD.DATA, "JUSD" and "DATA" are qualifiers.  The
         entire string is  considered to be a fully qualified
         name.   The name "JUSDDATATEST" on the other hand is
         an unqualified name since there are more than eight
         characters and no periods used as delimiters.
         Unqualified names are considered nonstandard at

     c.  The first character of the data set name and the
         first character following a period must be an
         alphabetic or national character.

     d.  The last character of a data set name may not be a
         period; nor may there be two consecutive periods.

NCC-IBM USER'S GUIDE                                     JAN  1988    NCC-IBM Conventions for User Data Sets

At the NCC,  additional restrictions are placed on the names of
data sets on public (USRxxx) and private disk volumes.  These
conventions  exist to ensure that one user's name does not
conflict with those of another and that storage charges are
billed and accounted for properly.  All data sets residing on
public or private disk storage are required to be catalogued and
to have standard names.  Uncatalogued data sets and those with
non-standard names on public disks will be deleted; those on
private disks will be archived with a retention period of two
years.  Non-standard names are defined as any name not having a
prefix of the form:


where iii represents the User-ID and aaaa represents the account
number. .No NCC restriction is placed on the remainder of the data
set name, but the data set must be catalogued; thus the name must
conform to the OS standards above and the cataloging restrictions
outlined below.

Given a name in the form above, the system can determine its
owner, the user whose User-ID is iii or a member of group aaaa.
A data set on a public disk volume is not automatically protected
against destructive access by another user unless explicitly
protected through the security system.  (Refer to IBM manual
"OS/VS2 RACF COMMAND LANGUAGE REFERENCE", number SC28-0733).    NCC-IBM Conventions for Naming VSAM Data Sets

When a VSAM data set is defined (must use IDCAMS), it must have a
cluster name and a data name.  KSDS also requires an index name,
and optionally an alternate index name and path name.  The
cluster name must follow current NCC standards as shown in
examples below, and the second level of the data set name must be
"VSAM".  The data name is formed by appending '.DATA1 to the
cluster name, while the index name is formed by appending
'.INDEX* to the cluster name.  The alternate index name is formed
by appending '.AIX' to the cluster name.  The path name is formed
by appending '.PATH1 to the cluster name.

An example follows:

     DEFINE CLUSTER(NAME(iiiaaaa.VSAM.xxxx.yyyy)...

NCC-IBM USER'S GUIDE                                     JAN 1988

     DEFINE ALTERNATEINDEX(NAME(iiiaaaa.VSAM.xxxx.yyyy . AIX) . . . )

     DEFINE PATM(NAME(iiiaaaa.VSAM.xxxx.yyyy.PATH). . . )

If the VSAM data set is to be associated with a group account
number, the following group data set naming convention should be
used :

     DEFINE CLUSTER(NAMECaaaa.VSAM.xxxx.yyyy)...

CICS VSAM data sets will follow these standards except that the
last index on the cluster will be assigned to  the CICS System

These  standards are enforced through a user exit.  Data sets not
meeting these standards will be deleted.

VSAM data sets must also reside on volumes dedicated to this
purpose.   Users should contact User Support (see Appendix A for
telephone number) to receive an allocation on the proper volume.
VSAM data sets residing on unauthorized volumes will be deleted.

3.3.1.M    NCC-IBM Conventions for System Data Sets

Library names for vendor-supported software are standardized
under  the following naming conventions =

     a.  For IBM software, the prefix is SYS1.

     b.  Vendor packages are prefixed with SYS2.prodname.
         where prodname describes the vendor package;  e.g.,

     c.  Documentation is prefixed with JUSD.

Other  qualifiers such as LOAD and OBJECT, are  added from a
standardized list for descriptive purposes.
                              3-1 1

NCC-IBM USER'S GUIDE                                     JAN  1988    System Catalog

The Operating System maintains a special data set, the System
Catalog, which contains data set names with pointers to the
appropriate unit type and specific volume(s). By referencing the
catalog, the operating system can locate any stored data set
without the UNIT and VOLUME parameters in a user's Job Control
Language.  This allows a user to refer to a data set only by its
name and to move a data set from one volume or storage medium to
another without changing any JCL streams or catalogued
procedures.   Because the catalog allows only one occurrence of a
given data set name, name conflicts cause a warning to be issued.

An entry is made in the catalog for each qualifier in a fully
qualified name with pointers to the lower-level qualifiers or to
the qualified name. With this hierarchical organization, certain
utility programs can list all the data sets with any specified
level of qualification. Because of this structure, a restriction
is placed on data set naming as follows:  a character string
which is used as a qualifier may not also be used as a qualified
name (i.e. the last qualifier in the data set name) when both are
preceded by an identical series of qualifiers.  For example, if
one were faced with the prospect of naming a program source
library for a special project and another program source library
for general use, one could not name one of the libraries:


and the other


since in the first case SOURCE is just a qualifier and in the
second it establishes the fully qualified data set name. The
catalog cannot resolve both.  However, if the first were named
iiiaaaa.SPROJECT.SOURCE, no conflict occurs with the second.

Catalogued data sets whose names do not conform to the
conventions described above are subject to being deleted at any
time Users who need to catalog such data sets should contact the
NCC-IBM User Support Department (see Appendix A for telephone

NCC-IBM USER'S GUIDE                                     JAN  1988

3.3.2    Job Scheduling    JES2

The RJE facility allows remote workstations to use the
capabilities of the system even though they are not located at
the central installation.   These remote workstations communicate
with the central computer  by means of an access method (for
example, TCAM or VTAM).  JES2 controls the remote unit record
devices as if they were local readers, printers, and punches.
The system can simultaneously operate local devices, read jobs
from remote readers into the queues to await processing, and
print and punch the results of processing at the remote site.

JES2 is a job entry subsystem which has assumed many of the
functions formerly performed by the job schedulers and HASP II
under previous releases of the Operating System (OS/MVT, and
OS/VS2 Release 1).  These  include:

     o   Receiving jobs from local devices, from system
         tasks initialized at the main console, and from
         remote locations.

     o   Maintaining all data submitted with jobs.

     o   Controlling initiators.

     o   Validating JCL before job processing.

     o   Job scheduling.

     o   SYSOUT processing.

     o   Routing of SYSOUT data sets to remote devices.

     o   Restarting jobs after system restarts.

JES2 commands and control  statements allow the programmer a
degree of control over the submission, execution, and retrieval
of a job.  The format of the card used for control statements is=

    col  1 3
         /^command operands

Refer to Appendix C for details of JES2 control statements.

JES2 operator commands are entered into the system from an RJE
station.  The commands are submitted through the card reader or
the operator's console, if available.  Output for these commands
is listed on the RJE printer or console, if there is one.  The
format of the command is as follows:

NCC-IBM USER'S GUIDE                                     JAN  1988

    col  1   4
         /*$command operands

The "/*" is omitted if the command is entered through the

Refer to Appendix D for details of JES2 operator commands.

All batch jobs which are executed on the NCC-IBM run under the
control of one of the active initiators.  Each initiator is set
up to handle one or more different job classes.  A job class  is
simply a queue for similar jobs.  At the NCC, job classes are
assigned automatically by the Job Stream Manager (JSM), a system
software module designed for the NCC-IBM.  Job classes are
defined according to the CPU time a job requests.  The majority
of jobs submitted by users are assigned to one of three job
classes which are defined as follows =

    Class I   - jobs requesting 3 CPU seconds or less.

    Class II  - jobs requesting 15 CPU seconds or less.

    Class III - jobs requesting 30 CPU seconds or less.

    Class IV  - jobs requesting 120 CPU seconds or less,
                but more than 30 CPU seconds.

Within job classes, jobs are ordered by the initiation priority.
The priority establishes two things:  the job's importance
relative to all of the other jobs which are assigned to the same
job class,  and how much the user is willing to pay for job
turnaround.  The service goals and cost multipliers are shown in
Figure 3-1  for each job initiation priority.  Note that priority
2 is the default job initiation priority.  Although NCC's
established service goals are to always process priority 1 jobs
overnight,  there is no guarantee that this will be the case.
Hardware and software problems as well as a workload too great
for the system to process can drastically affect turnaround
times.  Additionally, Production Services maintains a
concentrated effort to ensure that stability and performance
goals of no more than 8 failures per month, a Mean Time Between
Failures (MTBF) of 90 hours, and 99 percent uptime are met or

The user should note that the JES "priority" (7, 8, etc.) seen
for a job is a system-assigned priority using your requested
priority and other system considerations.

                                     JAN  1988

     2 (default)
       Class I
       Class II
       Class III
       Class IV

     1  (all
Service Time

Better than 3
Better than 2

90% of all jobs
done in < 10
90% of all jobs
done in < 30
90% of all jobs
done in < 60
90% of all jobs
done in < M hours

See specific
priority above.
Cost Multiplier

6 . 0 times normal
2.0 times normal
1.5 times normal

1 . 0 times normal
1 . 0 times normal

1 . 0 times normal

1 . 0 times normal

 .5 times normal

  Weekday rates
           Figure 3-1.  Job Service Level Definitions

A job submitted at priority 5 receives the fastest batch
turnaround available by reducing the start queue time.  It
normally will start within five minutes.  These jobs are charged
with a cost multiplier of six.

Users requiring priority 5 (urgent) job processing must
communicate that need to their respective ADP Coordinator.  The
ADP Coordinator is responsible for the budget of the project, and
since priority 5 is relatively expensive, the Coordinator must
approve this request.  By submitting a notification in writing to
the TSSMS Office (see Appendix A for address and telephone
number), an ADP Coordinator may delegate authority to another
person(s) who may then approve priority 5 requests for his

The ADP Coordinator or his authorized representative may request
priority 5 job processing by calling Data Processing Support
Services (see Appendix A for telephone number) and providing
name, account code, and job name.  DPSS then verifies this
request.  The request must be made prior to submission of the job
into the IBM system in order to receive the priority 5
classification.  The system checks the Job Stream Manager upon
submittal of the priority 5 job.  If the JSM has not been alerted
by DPSS that the job is authorized to run "urgent", then the

NCC-IBM USER'S GUIDE                                     JAN  1988

system reduces the priority to M.

The Operations Department is not authorized to start any job  out
of its normal sequence unless it conforms to these requirements.

In summary, the priority/class scheme acts as follows:

     o   Time/Region requirements limit service levels.

     o   The default service level is priority 2.

     o   Jobs will be assigned the service levels for which
         they qualify, not the service level requested.

In addition to the priority considerations mentioned above, there
are several other factors which can affect the timely processing
of a job.  These factors are =

     o   JES2 control statements /^BEFORE and /*AFTER.
         These statements are covered in Jippendix C.  Their
         purpose is to force jobs to run in a predefined
         sequence or after a specified time.

     o   JES2 control statement /*CNTL with the control
         statement SHR or EXC (see "SHARED Control
         Statement" or "EXCLUSIVE Control Statement" in
         Appendix C).  Their purpose is to denote that
         competing jobs may or may not share a computer

     o   The disposition of the data sets called for in the
         job's JCL.  For instance, if a job uses a data set
         called iiiaaaa.name and refers to it as DISP=OLD
         then (1) no other job will be al.'.a to use
         iiiaaaa.name until this job has released the data
         set and (2) this job will have to wait until all
         other jobs that are using iiiaaaa.name have
         released it.  A JES2 user modification has altered
         the normal JES2/MVS disposition processing flow.
         Now, a JES2 job selection exit issues a test
         ENgueue against the job's data sets.  If conflicts
         are detected, the job is not selected for execution
         and the TSO user is notified.  There are several
         circumstances where the exit will infrequently
         select the job to execute and ENfiueue conflicts
         will occur (GDG's, DSN refer backs, large number of
         data sets, timing).  The intent of the exit is to
         not select a job for execution if a data set EN2
         conflict will occur and thereby prevent initiators
         from being tied up with jobs waiting on data sets.
         Disposition and scheduling matters related to data

NCC-IBM USER'S GUIDE                                     JAN  1988

         sets are discussed under the DISP parameter in  the
         IBM publication OS/VS2 MVS JCL Reference Manual
         (Order No. GC28-1300) or MVS/XA JCL Reference
         Manual (Order No. GC28-11M8).

     o   Tape conflicts.  When more than one job requests
         access to data sets that reside on the same tape
         volume(s), timely JES2 processing may be impacted.

     o   The availability of hardware resources.  Whenever a
         job is selected for initiation, the resources that
         it requires are requested of the operating system.
         Main memory is sought first, then devices and/or
         data sets in the order they were specified in the
         JCL.  Whenever a resource request fails, the job is
         requeued and the next highest priority job will be
         tried.  (Users can influence this situation by  the
         use of the DEFER and AFFINITY subparameters of  the
         UNIT parameter on a DD statement.)  In MVS, the
        . upper limit of core that can be allocated to a  job
         is specified by the REGION.  Core is not allocated
         until it is needed.    User Production

The /*AFTER jobname, start-time JES2 command allows the user to
perform regularly scheduled production work on a recurring basis
without human intervention.  This method is attractive because it
reduces the need for daily human submittal of production
processing.  The following discussion is based on experience with
the /*AFTER statement in User Support.

The /^AFTER statement can be used to schedule a job to begin
execution after some specific time.  (Note:  This does not imply
that the job will begin immediately after the specified time.)
Because the /*AFTER statement has a restriction of 20 hours built
into it, a job may not submit itself for later execution on a
recurring 24-hour basis.  However, this restriction is easily
surmounted by using a two-job process.  One job submits work for
late at night, and the late night job submits an early morning
job for execution.  Each of these jobs in turn submits other jobs
to the internal reader for execution.  Two examples are given
below.                                   •;.. •

Example of Morning Job

     //uuuAM!   JOB (aaaafimas,Bnnn),'user-name',TIME=(,3),PRTY=4
     /*AFTER 0600

NCC-IBM USER'S GUIDE                                       JAN  1988

     //*        CNTL(STARTAM)
     //SYSIN    DD DUMMY
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB1)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB3)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BIKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBM)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BIKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBS)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB6)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB?)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
     //SYSIN    DD DUMMY
     //SYSUT1   DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBS)
     //SYSUT2    DD     SYSOUT=(A,INTRDR),
     //     DCB=(BIKSIZE=80,RECFM=F,BUFNO=1)

                                         JAN  1988
     //SYSIN    DD DUMMY
Example of Late Night Job
JOB (aaaafimas,Bnnn),'user-name',TIME=(,3) ,


The morning job submits the night time job, and the night time
job submits the morning job.  This is an infinite loop, so the
user is advised to be careful.  Users who have jobs requiring a
variation on the basis of the day of the week should refer to the
program JUSD.VSFORT2.DATA(FORDAT2) which prints out a variety of
date and time formats including the day of the week.  This
mechanism works even if the machine is brought down for scheduled
maintenance, as it is on Sunday evenings at the NCC.

The morning job submits the STARTPM job first.  STARTPM contains
a /*AFTER 2300 statement which causes the job to wait until  11 PM
even though it is a priority M job of 3 seconds maximum CPU  time.
JOB2 is also submitted by STARTAM.  JOB2 has a /*AFTER STARTPM
statement which causes it to wait until after the STARTPM job has
finished processing.  This will occur shortly after 11 PM since

NCC-IBM USER'S GUIDE                                     JAN  1988

STARTPM is a small (3-second), high priority (M) job.  Each of
the other jobs submitted by the morning job is chained to the
preceding job.  JOBn contains a /*AFTER JOB(n-l) statement which
causes it to wait until after the preceding job has finished.

This scheduling method has several advantages which combine to
yield reliable and timely execution at tl... NCC.

     o  Small CPU times.  A long job which might be further
        delayed because of a requirement for many minutes of
        CPU time is broken down into a series of smaller
        jobs.   These smaller jobs will tend to be processed
        ahead of longer running jobs.

     o  Saving the intermediate results.  The small jobs
        also save their results so that in the event of some
        catastrophe, the chain need not be restarted at the
        beginning but at the point of failure.  Frequently
        this can result in significant savings.

     o  gueuing the jobs many hours before scheduled
        execution.  As small jobs that have been in the
        backlog for hours, these jobs tend to be high in the
        queue for their given class.

If you are interested in a "management by exception" process, you
can submit these jobs to the hold queue (on the JOB statement put
MSGCLASS=P), and then have a last, or cleanup, job which runs
TSOBATCH and uses the TSO OUTPUT command to either save the job
in a DASD data set or delete (purge) the job from the HOLD queue.

Note:  The MSGCLASS=P HOLD queue is not the same as the /*ROUTE

The last clean up job could then issue a status message to the
user informing him of any jobs that terminated in an unusual
3.3.3    Direct Access Storage Device (DASD) Management

The DASD Management function is performed by a staff within the
Production Services Department.  Its main functions include
ensuring integrity of user data on DASD, maintaining maximum
space availability on DASD, and promoting optimal use of DASD
within the Data Center.   ASM2 is the main software package used
to accomplish these functions.  FDR, DF/DSS, and FAVER are also
heavily used.

NCC-IBM USER'S GUIDE                                      JAN  1988    Archives

Data sets which are unused for a designated number of days  are
routinely archived to tape for 1 year.  The number of days  of
nonuse is determined by the DASD management staff and may be
lengthened or shortened depending upon DASD usage and availbility
to insure sufficient DASD space to meet user requirements.  With
the implementation of ASM2's Intelligent Transparent Restore
(IXR), users should not be impacted by the floating interval  of

When data sets are archived, a record for each data set is
written to the archive catalog and resides there until the  tape
expires.   All Data Management archive tapes are copied for  safety
with the duplicate tape stored offsite for disaster recovery.
After the duplicate is created, the data sets are deleted from
the disk to release space.

Users may request the archiving of data sets for which they want
long-term retention on tape.  Two types ^.1 archives are available
for this purpose:  Archive Explicit (for two year retention)  and
Archive Permanent (for seven year retention).

To archive a data set for two years (Archive Explicit), the user
should enter the following command through either batch or  ISO=

     *AR 'iiiaaaa.name'

To archive a data set for seven years (Archive Permanent),  the
user should enter the following command through either batch  or

     $AR 'iiiaaaa.name' PERM

It should be noted that while the commands identify specific  data
for archival, actual archiving of the data sets is done by  runs
executed by NCC Data Management personnel several times a week,
i.e., the data set is not archived at the time the command  is
issued.  It may be archived several days later.  Users are
charged only for issuing the command, not for the unload  or
storage of the data set.

'JUSD.UTILITY.DATA(ARCHIVE)' is available to users for archiving
data sets.

From TSO, users may enter HELP $AR for an on-line description of
the $AR command.

NCC-IBM USER'S GUIDE                                     JAN  1988    Intelligent Transparent Restore Facility

Intelligent Transparent Restore Facility is a software feature
that allows automatic, real-time processing of "reload"
operations for DASD data sets and VSA11 clusters archived by ASM2 .
Prior to IXR, such operations were normally requested voluntarily
by the user when it was discovered that needed data had been
archived.   Additionally, ASM2 normally queued such requests so
that they could be processed asynchronously in batches at
predetermined intervals.

In two ways, IXR represents a major change from the
voluntary/queued reload approach, which will continue to be
supported.  First, the recognition that needed data has been
archived and is absent from on-line DASD is completely automatic
for batch jobs only.  This permits batch jobs to be executed
freely without voluntary predetermination of the status of data
to be accessed.  Second, when IXR discovers that archived data is
going to be accessed, the processing to reload the data to DASD
is performed in real time while the accessing program waits.  IXR
is not activated for TSO sessions when an archived data set is

With IXR,  users do not need to verify that data sets used in
batch jobs are active on the system and not in archive status.
Batch jobs no longer abend with "data set not found" conditions
caused by data sets previously having been archived.  No
intervention by users is required to reload archived data sets
referenced in batch jobs.  They are reloaded automatically.   IXR
is implemented using JOB staging which means that all data sets
referenced in the entire JCL job stream are submitted to IXR  for
availability analysis and possible reloading.  Job staging takes
place before job initiation (before any : - b step programs have
been executed).

Batch users do not have to take any action to invoke IXR or
reload a data set from archive via IXR.  Presently, IXR will  not
function with CICS, but the standard ASJ12 commands such as $RA
will continue to be supported so that any data set can be
reloaded from archive by the user's issuance of the appropriate
ASM2 commands.

IXR is implemented with two restrictions.  Once a data set has
been reloaded with IXR, it cannot be reloaded again with IXR
until it has been archived either by the user or by NCC Data
Management due to nonuse.  It can, however, be reloaded
explicitly by using the ASM2 $RA command.  IXR is not invoked for
Generation Data Group data sets to avoid corruption of the
cycling limits defined by the user.

IXR messages are available through the SYSABEND CLIST.  Users can

NCC-IBM USER'S GUIDE                                     JAN  1988

view these messages by keying in SYSABEND after the ISO ready
prompt and entering an IXR code when prompted, i.e., IXR0300A,
IXR0310I, etc.  A description of the IXR code will then be
displayed on the user's terminal.

There are rare instances when a user may want to turn off IXR for
a batch job.  Although not encouraged, this can be done by
inserting the following DD statement after the first EXEC
statement in the JCL=

        //XSJIXRN DD DUMMY    Reload From Archives

Users may observe information about their archived data sets by
entering the command:

     $AI da('iiiaaaa.qualifiers') ALL  [XONO] [CYCLE(n)] [PERM]

where up to five data set name levels may be specified either as
a mask or the complete data set name.  Other parameter
explanations are as follows =

     ALL      - Archive list will include archived data sets which
                were previously restored to disk (restoring the
                data set does not remove it from the archive catalog).
                Those that have been restored will be flagged with
                an "R".

     MONO     - Miscellaneous heading information will not be displayed
                in the archive list; use of NONO is recommended.

     CYCLE(n) - Inquiry will be limited to "n" days into the past
                (saves time and costs).

     PERM     - Archive list will be restricted to data sets residing
                in permanent archives.

A single data set may have several versions stored in the archive
catalog.  Any or all of the existing versions may be restored;
however, after one version is restored, it must be deleted or
renamed before the next version is restored.  Multiple data sets
by the same name may be stored in the archive catalog but may not
be active on the system.

'JUSD.UTILITY.DATA(ARCHLIST)' is available to users for listing
their archived data sets.   From TSO, users may enter HELP $AI for
an on-line description of the $AI command.

NCC-IBM USER'S GUIDE                                     JAN  1988

The user should ensure that the data set he wants to restore  does
not exist on disk before submitting the $RA command.  If it
already exists, the restore will not be successful.  Batch users
should rarely use the *RA command due to the automatic data set
restore facility now available (see Section on IXR).

Archived data sets may be queued to be reloaded by the user
through ISO (or batch, if necessary) using the following command:

     *RA 'iiiaaaa.qualifiers'[(-m)]

where the data set name is as it appears in the archive list.
The version "m" levels previous to the most recent will be
reloaded.  If "(-m)" is not given or if "m" is zero, the most
recently archived version of the data set will be restored.

ISO users pay only for queuing the reload command, not the actual
reload.  If a batch job is used for reload, all $RA commands
should be placed in the same job.

Users should beware of the fact that when Generation Data Group's
(GDG's) are restored, they always come back uncataloged.  The
reason for this is that when a GDG is archived, it is removed
from the disk, thus making its generation number available again
for future generations.  When a user restores a GDG, he should
catalog it by specifying the volume serial stated in the MAIL
data set (see Section for HAIL).  If the user does not do
so, the GDG will be scratched in the next day's scratch run.
Users have the responsibility for cataloging restored GDG's.

All other data sets are cataloged upon restoration, if it is
possible for them to be cataloged.

•JUSD.UTILITY.DATA(DEARC)' is available to users for queuing  data
sets to be restored from archives.  From TSO, users may enter
HELP $RA for an on-line description of the $RA command.  Data
sets are restored by BRELOAD, a started task that is constantly
active.    Incremental Backups

Incrementals are the nightly saves done on data sets new or
changed on the DASD since the last incremental was run.  Each
nightly save is retained for 31 days, at which time the tapes are
scratched.  Only data sets which are new or changed are saved in
incrementals;  those which are used only are not backed up.  A
record of each data set is maintained in the backup catalog until
the tape expires.

Users may observe  information about their backup data sets by

NCC-IBM USER'S GUIDE                                     JAN  1988

entering =

     $BI da('iiiaaaa.qualifiers') ALL  [NONO] [CYCLE(n)] [PERM]

where up to five data set name levels may be specified either as
a mask or  the complete data set name.  Other parameter meanings
are as follows =

     ALL      -  Backup list will include data sets which have
                previously been restored (restoring the data  set
                does not remove it from the incremental backup
                catalog).  Those which have been restored will be
                flagged with an "R".

     NONO      -  Miscellaneous heading information will not be
                displayed in the backup list; use of NONO is

     CYCLE(n) -  Backup list will be limited to "n" days into
                the past (saves time and costs).

A single data set may have several versions stored in the
incremental catalog.  Any or all of the existing versions may be
restored;  however, after one version is restored, it must be
renamed before the next version is restored.  Multiple data sets
by the same name may be stored in the backup catalog but may  not
exist on the system.

'JUSD.UTILITY.DATA(BKUPLIST)' is available to users for listing
their backed up  data sets.  From TSO, users may enter HELP $BI
for an on-line description of the $BI command.

The user should  ensure that the data set he wants to restore  does
not exist  on disk before submitting the $RB command.  If it
already exists,  the restore will not be successful.

Backed up  data sets may be reloaded by the user through TSO (or
batch, if  necessary, using the following command:

     *RB 'iiiaaaa.qualifiers'[(-m)]

where the  data set name is as it appears in the backup list.  The
version "m" levels previous to the most recent will be reloaded.
If "(-m)"  is not given or if "m" is zero, the most recently
backed up  version of the data set will be restored.  Data sets
are restored by  BRELOAD, a started task that is constantly

gueuing a  data set to be restored from incrementals may be done
through either batch or TSO.   TSO users pay only for queuing  the
reload command not the actual reload.  If a batch job is used,

NCC-IBM USER'S GUIDE                                     JAN 1988

all *RB commands should be placed in the same job.

'JUSD.UTILITY.DATA(RESTORE)'  is available to users for queuing
data sets to be restored from incrementals.   From ISO, users may
enter HELP $RB for an on-line description of the $RB command.    fiRELOAD's

When users submit $RA or $RB commands, the data set name go into
a queue backlog for reload.  The automatic BRELOAD function which
performs the restore runs periodically all day long.  If too many
commands are in the queue, BRELOAD only does a certain amount for
any given time.  For this reason, it is recommended that data
sets be reloaded in groups of 20 or less.  A large number of
reload commands submitted at one time uill slow down the reload
process for all users.

When a data set has been reloaded, either successfully or
unsuccessfully, the User-ID and account or Group-ID of the person
who submitted the request is notified via MAIL; i.e., a message
is put into iiiaaaa.MAIL.   If the message states that the data
set was not successfully restored, the user should not
necessarily resubmit the command.  DASD Management personnel
monitor the BRELOAD's all day long, and will resubmit failed *RA
and $RB commands under  the following conditions =

     o   Not enough space on the original volume to restore
         the data set.

     o   Uncataloged version of the data set exists due to
         an abnormally  terminated BRELOAD.

     o   BRELOAD processor abended.

     o   Any other problem not directly caused by the user.

DASD Management will not resubmit $RA and $RB commands for data
set restores which failed for one of the following reasons:

     o   Cataloged version exists on DASD.

     o   Data set not in incremental backup catalog (for
         $RB) or not in archive catalog (for $RA).

If the user gets an unsuccessfully restored message in his MAIL
data set, he should first check to see if the data set in
question is not already cataloged.  Then he should check to see
if it is in the backup  or archive catalog.  (The data set name
may be incorrectly keyed in.)  If the user still cannot figure
out the reason for the  unsuccessful restore, he should call User

NCC-IBM USER'S GUIDE                                     JAN  1988


If a user has a large number of data sets to restore, he should
submit a small group such as 20 or less, wait for the successful/
unsuccessfuly restore messages to appear in his MAIL data set,
then submit another group of 20.    Full Volume Saves

Full volume saves are performed biweekly by Data Management
personnel.  These biweekly save tapes have a retention period of
4 weeks and are stored in a vault for disaster recovery purposes
until they are ready to expire.

Selectively restoring data sets from full volume dumps is no
longer supported except in emergency situations since the tapes
will be kept offsite.  Because the retention period has
decreased, there should only very rarely ba a need to restore
data sets from full volume backups.   However, if it is necessary,
a charge of $13.00 per data set will be applied.  The user should
contact User Support who in turn will notify Data Management.  It
is not possible for a user to restore a data set from a full
volume backup.  Twenty-four hours are required to complete the
restore.   The user should check to see if the data set in
question is in archives (which have a 1 year retention period) or
incremental (via $AI or $BI) before calling User Support with the
request.   If it is in the archive or incremental catalog, the
user should restore the data set himself by issuing a $RA or $RB
command or depend upon IXR to restore it via a batch job.    Scratch

Certain types of data sets are routinely scratched from the user
disk packs.  The following types of data sets are scratched

    o  Uncataloged and have a creation date not equal to
       today's date.

    o  Containing no organization and are unused for one day.

    o  Empty and unused for seven days.

    o  fiRELOAD MAIL data sets which are over one week old.

    o  Nonstandard-named.

NCC-IBM USER'S GUIDE                                     JAN  1988    Compress

Once during the week, partitioned data sets which have an
allocation of greater than ten tracks are compressed.  Because of
the length of time involved to run this job, it is set up to
compress the partitioned data sets on a subset of user packs.
Thus a user's data set will be compressed by DASD Management
approximately once per month.

If the user finds it necessary to compress his data sets more
often, he may do so by the available means.
•JUSD.UTILITY.DATA(COMPRESS)' is available for this purpose.
From TSO, users may enter HELP *SM for an on-line description of
compressing a data set.  Also, under ISPF menu options 3.1 or
FILEAID 3.1, a command option of C uill compress the partitioned
data set whose name is entered in the data set name field.    Space Release

A nightly release job is run which releases all unused space in
sequential data sets that meet specific criteria.  It also
releases 90 percent of the unused space in partitioned data sets
that meet specific criteria.  The purpose of this is to release
space in over-allocated data sets and to increase available space
on the system.

On a regular basis, the automated release of unused space
allocated to Direct Access (DA) data sets- is performed during
non-prime shift using the following criteria:

     o  Direct access data set with at least an 80 percent
        utilization of allocated space will not be subjected
        to space release.

     o  Direct access data sets created within the previous 30 days
        will not be subjected to space release.

     o  Space for all other direct access data sets will be main-
        tained at no more than the number of tracks used plus
        30 percent of that total.

Two members in ' JUSD.UTILITY.DATA' are available to users for
space information.  SPACEMAN performs space management functions;
SPACERPT lists the space allocated and used by data sets.    Acquiring Disk Space

Disk space on public volumes may be acquired for permanent data
set storage by using the proper TSO commands or by submitting a


NCC-IBM USER'S GUIDE                                     JAN  1988

batch job with the proper JCL.   See Section 3.3.1 for naming  and
cataloging requirements.

To create a data set on a public volume from a batch job, the
user must include a DD card in the JCL stream having the
following parameters:

    o  UNIT=DISK (Disk represents any of a group of direct access
       devices), or UNIT=SYSDA.

    o  DSNAME

    o  DISP of (NEW,CATALOG,DELETE) - the DELETE option should always
        or     (MOD,CATALOG,DELETE)   be used because it deletes
                                      garbage data sets created by
                                      jobs which abend.

    o  Appropriate space  and DCB parameters.

For information on how to determine space and DCB parameters
appropriate for the STC 8380 DASD at the NCC, users are
encouraged to reference Section M . 1 . 3 . 1 on disk data sets and
Section M.1.3.2 on allocating DASD space.

All data sets must be  catalogued by the user at creation or they
will be deleted in the next day's scratch run.  Creating a data
set on disk prior to its  first use is seldom necessary.  Space is
automatically allocated by the operating system when a
disposition of NEW and the appropriate SPACE parameters are coded
in the DD statement that  creates the data set.  Users should  use
a DISP of OLD or SHR if they want to write over a currently
existing data set.

Users may not specify  volume serial numbers for user packs when
allocating a new data  set on DASD.   The operating system will
allocate the data set  on  a pack with sufficient space.  NCC DASD
Management often moves data sets to provide adequate space, and
if the user specifies  a volume serial for a data set which has
been moved, the job will  terminate.

After a data set is allocated on a DASD device, it is accessible
through TSO, SPF, and  batch jobs.  When accessing data sets
through batch, UNIT=DISK  or UNIT=SYSDA should be specified.
These are installation parameters which represent any of a group
of direct access devices.

After a data set is allocated on a DASD device, it is subject to
all NCC guidelines which  are delineated in this section.  All
data sets on user packs must have standard names and be
cataloged.  Other restrictions  are discussed in previous sections
of this manual.

NCC-IBM USER'S GUIDE                                     JAN 1988

Three members in 'JUSD.UTILITY.DATA'  are available to users for
determining information about data sets.  FIIEUSE lists the last
day a data set was  used and modified; DSNLIST and LISTCAT list
users' data sets.  From ISO, users may enter HELP $US for an on-
line description of the $US command which gives usage data for a
data set.    Temporary On-Line Data Set

There are two types of temporary disk data sets at the NCC.  The
first consists of data sets given a permanent name but intended
for temporary use.  These are generally written on system
reserved storage.  The second is given a temporary name and
intended for use within a single job.

Such a data set may be created in a user's JCL by specifying:

    o  UNIT=SYSDA.

    o  No VOL=SER parameter.

    o  A DSNAME of the form ££name where name is one to eight
       alphanumeric or national characters beginning with an
       alphabetic or national character.

    o  No DISP parameter or a DISP of (NEW,PASS) or (NEW,DELETE).

    o  Appropriate SPACE and DCB parameters.

The operating system will assign a fully qualified data set name.
This name begins with SYS and incorporates the time and date of
creation in addition to the user specific time.

If it is necessary to refer to a temporary data set in subsequent
job steps, the JCL need only contain a DSNAME parameter
specifying $$name and an appropriate DISP parameter.
Furthermore, the JCL cannot specify VOL=SER=, UNIT=, or VOL=REF=.
A DISP of (NEW,PASS) must have been specified in the step which
created the data set.  Temporary data sets will be scratched
after completion of the job for which they were created.    Private Disk Packs

Private disk packs may be obtained only if justification can show
that they are necessary.  Private disk packs at the NCC are
highly discouraged and are seldom given.   Justification must be
made in writing to the EPA Technical Manager for DASD management.
The following information should be supplied when requesting a
private pack=

NCC-IBM USER'S GUIDE                                     JAN 1988

     o   Name and User-ID of volume owner and an alternate

     o   Account code to be charged.

     o   Address and telephone number of owner and alternate

     o   The volume serial number(s) of the desired
         volume(s ) .

     o   Size and location of the Volume Table of Contents
         (VTOC) .

     o   Backup  schedule required by the user (incrementals,
         biweekly),  if any.

     o   Explicit need for the private pack.

Costs of private packs are discussed in Section 2.M.S.2.

Owners of private packs are responsible for efficient and
appropriate usage of it.  A private pack will be reclaimed by NCC
if it is not meeting the following usage criteria;

     o   At least 70% of the space must be used as well as

     o   At least 60% of the data sets must be accessed

     o   Total unused space on all private packs belonging
         to an organization must not exceed one pack.

Data sets on private packs should be referenced by their
cataloged name with no VOL=SER except at the time of creation.
The following restrictions are placed on data sets on private

     o   Cataloged  data sets left empty for 35 days are
         deleted weekly.

     o   Data sets  which are uncataloged are deleted weekly.

     o   Non-standard named data sets are deleted weekly.

     o   Data sets  which are unused for 180 days are
         archived for two years.

Owner backup of  private packs are discouraged, but if a user's
requirements are such that a backup is necessary outside of DASD

NCC-IBM USER'S GUIDE                                     JAN  1988

management's schedule, the responsibility for backup must be
borne by the user.  Users who find it necessary to backup their
own data sets and disks should contact Data Management for
further considerations.    Efficient Use of DASD at the NCC

To promote optimal use of DASD at the NCC, a description of some
common areas of DASD misuse are provided followed by suggestions
on how to avoid such abuses by using these storage devices
effectively and economically.  With user cooperation, misuse of
public on-line disk storage can be prevented and availability of
free space will increase.

The major areas of DASD abuse at the NCC are as follows:

     o  Improper blocking of data sets on STC 8380 DASD.

     o  Overallocation of space for data sets.

     o  "Touching" data sets periodically to circumvent the offload
        criteria for inactive data sets.

     o  Mass reloading of offloaded data sets even when they will
        not be used.

     o  Failure to use partitioned data sets to organize small data
        sets into libraries.

Improper blocking of data sets on STC 8380 DASD is often due to a
lack of knowledge.  As block sizes decrease, the amount of data
that can be stored on a track also decreases.  This is because
smaller block sizes increase the amount of space required for
overhead.  (There are more addresss areas and gaps.)  Users
frequently allocate a data set with a block size of 80 bytes and
other very small increments,  a practice that uses only as little
as 14 percent of the allocated space with the remainder consumed
by overhead.

The sizes of logical records to be blocked are another
consideration of block sizes.  A block size must be a multiple of
the length of fixed-length logical records.   One procedure for
arriving at a suitable block size is to start by selecting a
likely candidate as an ideal block size and then adjusting it to
accommodate the logical records in the data set.  For example, if
the data set contains 80-byte logical records, an ideal block
size of 6233 might be chosen.  The closest to this ideal block
size can be attained with 77 logical records blocked in 6160-byte
sizes.  This permits 7 blocks to be placed on a track and
achieves 90 percent utilization.

NCC-IBM USER'S GUIDE                                     JAN 1988

For more detailed information on allocating space by cylinder,
track, or bytes per block, see Sections 1.1. 3.1 and M.1.3.2 and
the following on-line data set and data set members =

     JUSD.DASD.BLKSIZE (discusses allocating DASD space)
     JUSD.STC8380. DATA(BLOKSIZE) (discusses use of the 8380 disk pack)

Many users  routinely oveallocate data sets rather than specify a
reasonable  size for the primary space allocation and include a
secondary space allocation to allow the data set to be expanded.
The primary space is allocated in its entirety when the data set
is created.  The secondary is allocated only after the primary is
exceeded.  Because the secondary provides a reserve for growth
potential over time, the primary amount can be conservative.  See
Section M.1.3.2 for guidance in allocation of space.

"Touching"  of data sets periodically to retain inactive data sets
on-line is  widespread among NCC users.  This practice is
unnecessary since NCC extended the mandatory archive retention
period to 1 year and installed IXR.  IXR reloads an archived data
set automatically when referenced by a batch job.  When a TSO
session references an archived data set, IXR is not activated.
IXR not only reloads data sets archived due to nonuse but will
reload data sets archived by the user for 2-7 years.   Therefore,
it is unnecessary to periodically "touch" data sets to circumvent
the offload criteria.  Those who persist in disregarding the
NCC's offload policies will be notified of their misuse of
resources.   Refusal to abandon the practice could lead to
limitations on resource usage.

Many users  unnecessarily reload masses of archived data sets even
when they will not be used.  Again, this practice is unnecessary
since IXR will automatically reload archived data sets.  As with
the practice of simply "touching" on-line data sets to keep them
from being  archived, those who abuse this capability will be
notified of their misuse of resources.  Refusal to abandon the
practice could lead to limitations on resource usage.

Using partitioned data sets to organize small data sets into
program libraries,  data libraries, JCL libraries, etc. benefits
the user and the entire NCC user community.  Such organization of
small data  sets into partitioned data set members increases
available DASD space, reduces overhead, and makes it easier for
users to manage and track their data sets. For example, the NCC
recently copied 50 one-track sequential data sets into a
partitioned data set for a user, thereby reducing the number of
tracks required by 78 percent.  Moreover, if all the user's 298
one-track sequential data sets had been organized as one or more
partitioned data ses, about 233 tracks would have been available
for other users.

NCC-IBM USER'S GUIDE                                     JAN 1988

Sequential data sets can be easily merged into a library using
ISPF.  Use ISPF to create the partitioned data set that will
serve as the library.  A partitioned data set must have space
allocated for the directory that will contain the member names
and pointers to the members in the partitioned data set.  The
directory is allocated in units of 256-byte records.  The number
of member references that can be stored in a directory depends
upon the length of the user area which is used to store

On the average, a value of five entries per directory block is a
good working estimate to use when determining the number of
directory blocks to specify.  For example, if you want the
maximum number of members to be 36, then allocate the blocks as
follows =

     CEILINGC36/5) = CEILINGC7.2) = 8 directory blocks

The CEILING of a value is the least integer greater than or equal
to the value.

Use ISPF 3.3 to copy individual sequential data sets into the
library.  Under ISPF 3.3 it is advisable to use the C command and
copy data sets rather than the M command to move data sets.   If
an error occurs, the copy will leave the source data set intact,
while a move could delete the source data set and result in the
loss of data.

In batch, directory blocks are specified in the SPACE parameter
on the DD statement.  The format of the SPACE parameter with the
directory specification is as follows;


In the above parameter, uppercase entries are written exactly as
specified; lowercase entries represent numerical values to be
specified by the user.

In batch mode, the IEBGENER utility can be used to merge
sequential data sets into a partitioned data set.  A sample job
step to copy a sequential data set named
XXXYYYY.MAY87,DAY18.TEST.DATA into the member D18TS of the
partitioned data set XXXYYYY.MAY87.DATA follows:

     //SYSIN       DD   DUMMY


NCC-IBM USER'S GUIDE                                     JAN 1988

The SYSUT1 DD statement specifies the input data set, and the
SYSUT2 DD statement specifies the output data set.  The job would
have to include a job step for each sequential data set to be
copied into the library.

After a successful copy of a sequential data set into a library,
delete the original sequential data set.

For further information on partitioning data sets, see the on-
line tutorial JUSD.DASD.PDS.
3 . 3 . M    Tape Management

3.3.M.1    Tape Data Storage

There are a number of tape drives at the NCC.   Each consists of a
spindle on uhich may be mounted a reel of tape (or tape volume),
a spindle on uhich an empty take-up reel is permanently mounted,
and a read/write mechanism between these spindles. The reels of
tape  mounted on these drives consist of a plastic or metal spool
wound with a strip of thin plastic tape (2400 feet is standard),
one-half inch wide and a few mils thick.  Distributed evenly on
this  tape is a magnetically sensitive layer of oxide, the medium
on which data is recorded.

In operation, the tape reel is mounted on the empty spindle. The
tape  is threaded past the read/write mechanism to the empty take-
up reel and positioned at a fixed starting point. Under program
control, records may be read or written, records may be passed
over  in either direction, and whole files may be skipped over in
either direction. Records are separated by gaps of approximately
.6 inches, and files are separated by larger tape-marks.

The read/write mechanism is capable of reading or writing data in
a number of tracks simultaneously. A track is considered to be a
one-bit-wide path parallel to the tape's length. At the NCC, all
tapes have either seven or nine tracks, one of which is used for
a character parity check, leaving either six or eight data bits
at any one position on the tape. Nine-track tapes with eight data
bits  are suitable for recording characters in the eight-bit
internal representation of the IBM System, or in ASCII.  Seven-
track tapes with six data bits per character are primarily
intended for recording data in a six-bit code which is standard
on many other computer systems.

In the longitudinal direction, the tape is divided into frames,
each  consisting of either six or eight data bits plus a parity
bit.  A single frame is one bit in length and either seven or nine
bits  wide. There are several standard densities which refer to
the number of bits (read frames) per inch. These densities are

NCC-IBM USER'S GUIDE                                     JAN  1988

given as a number followed by the abbreviation bpi, for bits  per
inch (e.g., 800 bpi). In general, the higher the density, the
greater the transmission rate and, perhaps surprisingly, the
greater the reliability of the device.

In order to read a particular record, all the tape between the
current position and the required record must be sequentially
passed under the read/write head. While the time between
consecutive records is very small, typically one to two
milliseconds, the delay between the first and last record on  a
tape reel could be as much as two minutes on even the fastest
tape drives. As a result, tape is used almost exclusively for
data files which are processed in their physical order, and which
need not be shared by more than one program at any given time.

Additional characteristics of tape data storage are:

     a.  Reels of tape are the most practical way to
         transport data from one physical location to
         another.  A reel of tape weighs one to two pounds
         and can accommodate 150 million characters or more.

     b.  When a data file on tape is to be altered, the
         entire reel must be copied to another reel with the
         appropriate changes made to individual records as
         they are carried forward.  This attribute makes a
         single reel of tape an awkward medium for the
         storage of volatile files.

     c.  A reel of tape may not be shared by multiple jobs,
         and two or more data sets on the same volume must
         be processed one after the other by a single job.

Tape Assignment

The NCC maintains a supply of 9-track 6250 bpi magnetic tapes for
general use.  These tapes are considered standard by the NCC;
however, the user can write to these tapes at 1600 bpi.  Other
scratch tapes available but considered nonstandard are as

     o    7-track, 200 bpi, labeled
     o    7-track, 556 bpi, labeled
     o    7-track, 800 bpi, labeled
     o    9-track, 800 bpi, labeled
     o    9-track, 1600 bpi, labeled
     o    nonlabeled, tape marked

These tapes can be accessed through the proper JCL statements and
options (see Section

NCC-IBM USER'S GUIDE                                     JAN  1988

The standard way to acquire magnetic tapes for use at the NCC is
to use the appropriate JCL in a computer run that creates a tape
data set.  Specifically, a DD statement uith the following
characteristics will result in an automatic assignment of a tape
volume from the pool of available tapes =

     a.    A UNIT parameter specifying a tape drive.

     b.    A DISP parameter of (NEW,KEEP) or (NEW,CATLG).

     c.    No volume identification.

When a tape is written as described above with DISP=(NEW,KEEP) or
(NEW,CATLG) the data set name and volume identification  are
written into the Tape Management Catalog (TMC).  Note that this
is not the system catalog discussed in Section

Computer tapes owned and supplied by the NCC are initialized and
prepared for use prior to job submission.  Tapes acquired from
external sources (foreign tapes) may require some preparation
before they can be used on the NCC-IBH computer system.  Such
tapes fall into two categories:   those written at some other
computer installation, and uninitialized tapes which have never
been used before.  The data set names and volume identifications
of tapes created at other installations are unknown to the Tape
Management System; they may even duplicate data sets or  volumes
already in the NCC Tape Library. Therefore, processing of such
tapes must bypass the Tape Management Catalog or access  will be
prevented by it.  Refer to Section for more details.

Uninitialized tapes create an even more bothersome problem when
used without proper initialization. A typical tape drive begins
reading forward from the beginning of a tape, looking for a label
or a tape mark as its very first operation. When no such label or
tape mark can be found (as with an uninitialized tape),  the tape
will usually run completely off the end of the reel. Therefore,
uninitialized tapes must be tape-marked or labeled before use.
DPSS will perform any needed preparation and initialization of
tapes upon request. In addition to labeling and tape-marking,
DPSS can also perform cleaning and/or degaussing of tapes as
requested. However, this process must be scheduled and will be
performed as promptly as possible, pending current workload.

Alien Tapes

Occasionally, an IBM, SPERRY, or DEC user must read an NCC tape
on an NCC computer system other than the one that created the
tape.  That tape then becomes alien to the system on which it is
to be read.  The unique tape numbering scheme for each NCC
computer system makes it necessary that the following procedures
be used before attempting to read an alien tape;

NCC-IBM USER'S GUIDE                                     JAN  1988

     a.  The owner of the tape must contact DPSS (see
         Appendix A for telephone number) to request that
         the tape be made available on another computer
         system.   DPSS will complete internal forms and
         verify that the caller is actually the owner.  If a
         job is submitted prior to contacting DPSS, the
         operator will cancel the job with an appropriate
         message to the user.

     b.  DPSS will place a file-protect ring in the tape to
         prevent any attempts to write on the tape.

     c.  The tape will be moved to the appropriate staging
         area for the computer system that is to read it.

     d.  At the end of the workday the file-protect ring
         will be removed, and the tape will be returned to
         its original tape library.  If, however, the tape
         is required on its native system prior to the end
         of the workday, the user should just submit the job
         that calls for the tape.

The NCC currently maintains over 55,000 tapes.  These procedures
are necessary for tracking, controlling, and preventing
unauthorized use of any of these tapes.

Retention Period

Upon tape creation, the user has the option to specify an
expiration date or retention period.  IMS allows a maximum
expiration date of 13 months.  If a retention period is not
specified, the system uses a default of five days from creation

Temporary Work Tapes

If a retention period is not specified and a temporary data set
name is used, either system-generated or explicitly defined
through the use of ££, the tape is considered a work tape.  The
tape remains in scratch status and may be used as such by
subsequent jobs.  A work tape may also be identified by specifying
RETPD=0 in the DD statement LABEL parameter. Specifying
(NEW,DELETE) as the DISP with no EXPDT or RETPD for the tape also
denotes a work tape (see Section M.1.3.3).

Recreating Tape Data Sets on the Same Volume

Recreation of the last (or only) data set on a tape volume is
allowed if the data set name specified is the same as the DSN
used to originally create the data set,  and the disposition
parameter is changed to OLD.

NCC-IBM USER'S GUIDE                                     JAN 1988

Write Ring Policy

It is standard procedure at the NCC for write rings to remain in
TMS-controlled library tapes.  For maximum data protection, users
may request that DPSS place a "NO RING EVER" external label on a
tape.  The request will be logged, and a file guard ring will be
inserted in the tape to disallow the insertion of a write ring.
Operators are not permitted to insert or remove file guards.  The
user must notify DPSS when he wants the file guard inserted or
removed.    Tape Management System

The Tape Management System (TMS) is an integral part of the NCC-
IBM operating system environment that provides the following
functional advantages =

     a.  Gives users additional capabilities that are not
         provided by the Operating System for handling tape
         data sets.

     b.  Reduces the human intervention required to handle
         tapes and operate a computer center.

     c.  Increases computer system throughput by reducing
         clerical load and manual operations required for
         handling tapes.

     d.  Provides additional protection against loss or
         misuse of tape data sets .

     e.  Provides useful management information, including
         daily logs of tapes approaching expiration.

The TMS software assumes most of the clerical burden associated
with tape handling.   Thus operators are relieved of accounting
and tracking chores related to tape processing, and users are
provided with the capability to treat tape files with much of the
ease usually associated with disk resident data sets.

A key component of the TMS is the Tape Management Catalog.  All
of the details about tapes under control of the TMS, including
volume identification,  resident DSNAME's, and expiration dates,
are recorded in the TMC.   It is this facility that enables the
TMS to locate and retrieve tape data sets without human

While standard OS conventions still apply to DD statements for
tape files, the Tape Management System's enhanced capabilities
require special considerations in the JCL as listed below for

NCC-IBM USER'S GUIDE                                     JAN  1988

tape data sets .

     a.  DSNAME - The data set name is the primary item of
         identification needed for the creation and
         subsequent use of a tape data set under TMS, even
         for unlabeled tapes.  The data set name is recorded
         in the Tape Management Catalog at the time it is
         created and is later compared to the DSN entry in
         the JCL when used for input.

     b.  VOL=SER - The volume identification must be
         supplied in the JCL, unless the data set is
         catalogued in the system catalog.  For uncatalogued
         data sets, the external volume-serial number must
         be specified, regardless of whether the tape is
         labeled or unlabeled.  The exception to this is
         that if the user is requesting a scratch volume, no
         VOL=SER parameter is given.  TMS will then assign a
         specific volume.

Warning:  if a job ABENDS, any scratch tapes allocated during the
run will be set to a retention period of four days (i.e.
EXPDT=98004) regardless of the original retention request.

     c.  LABEL - The retention period or expiration date
         subparameter of the LABEL parameter is used to form
         the TMS expiration date.  The TMS expiration date
         is either the EXPDT as coded, or the RETPD plus the
         creation date.  If no EXPDT or RETPD entries are
         given,  the system uses a default RETPD=5, yielding
         an expiration date five days frcu.. creation date .
         Therefore, users must remember to specify RETPD=nnn
         or EXPDT=yyddd to save the tape at the end of the
         job .

Several EXPDT subparameter values have special meanings.  All
values of 95000 and higher are reserved for special use under the

     a.  EXPDT=98000 indicates a tape volume for which there
         is no record in the TMC.  TMS processing is
         bypassed in this case.  For DISP=NEW tapes, the
         operator will verify that an external tape
         (foreign) is in fact being written.  If the
         requested volume is a TMS-controlled tape, the
         operator will cancel the job.

     b.  EXPDT=98ddd, where ddd is greater than zero and
         less than 366, indicates to TMS that the data set
         is to be scratched ddd days after the date of last
         use.   For example, given a tape with EXPDT=98001,

NCC-IBM USER'S GUIDE                                     JAN  1988

         the user can retain the tape indefinitely by
         referencing it at least once a day.  This feature
         allows retention of a data set until a certain
         period of nonuse has transpired, at which time it
         is scratched.

     c.  EXPDT=99000 indicates that the data set is a member
         of a Generation Data Group (GDG).  This value
         should only be specified with data sets which are
         created with a DISP of (NEW,CATLG), or else it
         provides no protection at all.  This value is to be
         used only for tape data sets which are members of
         Generation Data Groups (GDG's) so that when the
         data set is deleted from the GDG, the tape is
         released to the scratch pool.

     d.  EXPDT=99nnn, where nnn is an integer from 1 through
         364 specifies that the data set being created is
         one of a set of nnn data sets, each on a separate
         tape volume and all having the same data set name.
         TMS keeps track of the various versions of data
         sets like this in much the same fashion as the OS
         tracks the various generations of a generation data
         group.  However, this value should not be specified
         for data sets which are members of a GDG; its
         primary use is for data sets which are not
         catalogued in the system catalog.

TMS Inquiries

Information about tape volumes controlled by TMS can be obtained
via TSO or Batch either by volume number (VOL) or, if catalogued,
by data set name (DSN).

In TSO inquiries can be made by entering the command TMS.  The
system will prompt for a password.  If TMS is entered for the
password response, the system will prompt for a request or reply.
The six-digit volume number or a fully-qualified data set name
(entered without quotes) may be entered.  If a data set name  is
entered, it must be catalogued in the System Catalog.  If the
volume or data set is found in the Tape Management Catalog, the
IMC record will be displayed at the user's terminal.  If a record
is not found, the user will be informed and prompted for another
request or reply.  Another volume number ^/r data set name may
then be entered.

TMS inquiries are terminated by entering 'C1.

Batch inquiries on selected TMC records can be obtained by using
the following runstream:

NCC-IBM USER'S GUIDE                                     JAN 1988

     //stepname  EXEC TMSRE2,[PRINT=class]
     (control statements)

The following control statements are available:

     a.  VOL - requests a specific volume number or range of
         volume numbers.   If an inquiry is made to a record
         containing multiple data sets on a volume or
         multiple volume for a data set, all records
         pertaining to the set will be listed.  However, if
         the VOL control statement requests a range of
         volumes, only one record will be listed for each
         volume in the range.  The format is as follows:


         Where VOL is the control statement identifier and must
         begin in column one.

         Where vvvvvv is the volume serial number to be printed.
         If two volume serial numbers separated by a dash) are
         specified, one IMS record for each volume included in
         the range will be printed.

         Where format specifies the print format.  The values for
         format are =

         SHORT - the most important fields such as data set name
                 and DCB attributes are printed.

         LONG - all fields are printed (default).

         DUMP - record is printed in hexadecimal format.

     b.  DSN - requests a specific data set.  The data set
         must be catalogued in the System Catalog. The
         format is as follows:


         Where DSN is the control statement identifier and must
         begin in column one.

         Where data-set-name is the data set for which
         information is to be printed.  GDG's may be specified by
         specifying the data set name and the relative

         Where format specifies the print format (see a. above).

         Multiple control statements are allowed per execution

NCC-IBM USER'S GUIDE                                     OAN  1988

         and the TMSRPT DD statement may specify a DSN rather
         than SYSOUT.

The following example will print the short format of the IMS
volumes 000010, 000011, and 000012.  Volume 000015 will printed
in the long format.  The TMS volume containing the data set
iiiaaaa.z will be printed in the long format.

     DSN=iiiaaaa.z    Foreign Tapes

Foreign tapes are those not permanently stored in the tape
library nor resident in the Tape Management Catalog. This
includes any tape created with the intention of removing it from
NCC or any tape submitted from an outside source for processing
at the NCC.

In order to process a foreign tape, the user must code a special
value of 98000 for expiration date in the DD statement for that
tape which causes the Tape Management Catalog to be bypassed for
that volume.  Foreign tapes will never be part of the TMC but may
be copied to a TMS tape.  Tapes recognised as TMS tapes should
not be referenced using EXPDT=98000.

Foreign tapes provided by the user to be processed at the NCC
should be delivered or mailed to DPSS at the address given in
Appendix A.

Users are required by the NCC to properly identify their tapes
prior to submittal. Identification stickers should be applied by
the user to the face of the foreign tape, indicating:

     a.  User name.

     b.  User-ID.

     c.  Account number.

     d.  Telephone number.

     e.  Bin number or mailing address.

     f.  VOLSER or original reel identification.

Nonlabeled tapes must also be uniquely identified for cross-
reference during tape assignment by a job.

NCC-IBM USER'S GUIDE                                     JAN  1988

The user is responsible for the cost of shipping the tape to  the
NCC.   He will be notified of its arrival by DPSS.

Foreign tapes are considered ready for px-cessing as they arrive
at the NCC unless the user specifically requests DPSS to label
the tape, or to tape mark an unlabeled tape.  The user should
submit such a request to DPSS with the following information:

     a.  User's (requestor's) name.

     b.  Volume serial number with which the tape is to be
         labeled, if applicable.

     c.  Request to tape mark an unlabeled tape, if
         applicable.  Tape marking prevents the tape from
         running off the end of the reel upon first access
         and/or prior to creation of data on the tape.

When the user receives notification that the request has been
completed, he may process the tape.

NCC has established a system that uniquely identifies each
foreign tape. The Tape Librarian assigns a slot number to each
foreign tape, thereby establishing a unique identifier to help
the operator find the tape. The slot numbers indicate the system
on which the foreign tape is initially to be used.  However,
tapes can be used interchangeably from system to system, using
the same slot number originally assigned.  The slot number is of
the form:

     Bxnnnn where x = 0, 1, or 2, representing IBM, Sperry,
     and DEC, respectively; nnnn represents a number in the
     range 0001 - 9999.

No foreign tape will be placed into service until the user has
been contacted and given the slot number.

To use the foreign tape, the JCL tape DD statement must have  a
VOL=SER parameter to define the volume.  If the tape has a
standard label recognized by the IBM system -- LABEL=(,SL) or
LABEL=(,AL) (see Section M.1.3.3) — the VOL=SER parameter must
specify the same volume serial found in the internal label.   An
example is a tape brought from another installation which has
internal and external labels with serial number vvvvvv.  When
this tape is entered into the NCC-IBM as a foreign tape, it is
given a slot number (i.e. an external label) of Bxnnnn.  The  DD
statement must indicate VOL=SER=vvvvvv and EXPDT=98000 in the
LABEL field.  However, to insure the proper tape is mounted,  the
user must include the JCL tape mount statement as follows:

     /*TAPE Bxnnnn,vvvvvv

NCC-IBM USER'S GUIDE                                     JAN  1988

This will display a mount message to the operator for tape Bxnnnn
as well as allowing the system software to check the validity of
the tape's internal label against the volume serial number,
vvvvvv .

SPERRY tapes also have both regular serial numbers (starting  with
a numeric) and foreign tape numbers (of the form Blnnnn)
analogous to IMS-controlled and foreign tapes on the IBM.  If
processed on the IBM system, a SPERRY tape must always be used
with EXPDT=98000 in the LABEL field since it is "foreign" to  TMS.
The following sample of JCL may be used for SPERRY tapes.  If it
has a Blnnnn slot number, the /*TAPE card must be included;
otherwise that card should not be used.
                                        ' <.'
         /*TAPE Blnnnn,www
         //DDNAME DD VOL = SER = vwwv,
         // LABEL=(n,AL,EXPDT=98000),

Where n is the desired file sequence  and vwwv is the Sperry
VOLSER (e.g., VOL=SER=009127).   DCB=OPTCD=2 must be specified
when processing ANSI labels.

A file guard ring will be inserted on every foreign tape received
at NCC and a "no write ring" sticker  will be placed on the face
of the foreign volume.  This reversal of the former default
procedure will prevent inadvertent overwriting of the data on
foreign tapes.  Users who want to retain the ability to write on
foreign tapes submitted for processing at the NCC must notify
DPSS of such requirements.   This notification should be provided
at the time the tape is submitted to  NCC along with the foreign
tape number.  The notification must be complete before any job is
submitted that will attempt to write  on the tape.

Users should be aware that there is no security on a foreign
tape .

Within 90 days, foreign tapes may be  released by calling DPSS and
requesting the release. These tapes will be shipped upon request
at the NCC's expense.  After 90 days, such tapes will be
automatically released and sent to the person under whose User-ID
or name the tape is registered.  In order to permanently maintain
data sets or files stored on foreign  tapes at the NCC, such data
must be copied by the user to an internal NCC tape or disk.

NCC-IBM USER'S GUIDE                                     JAN  1988    Magnetic Tape Maintenance and Disposition

IMS-controlled tapes are intended to remain in the NCC Tape
Library or in the computer room.  Any tape created with the
intention of removing it from the NCC, either temporarily or
permanently, should be processed as a foreign tape, whether that
tape is provided by the user or purchased from DPSS .   NCC policy
states that IMS tapes sent offsite must be purchased (*15/each).
The slot will then be filled by another tape with the same

Some exceptions are made for the removal of TMS-controlled tapes
from the NCC for support for certain specific purposes, including
the following:

     o    Micrographics processing.

     o    Plotting at WIC.

     o    Keypunch processing.

     o    Archival storage.

     o    Payroll.

When a TMS-controlled tape is removed from or returned to the NCC
as in the cases above, DPSS  updates the TMS to reflect the

Monthly Foreign Tape Report

A monthly Tape Report is sent to the holder of every User-ID
having foreign tapes registered in the NCC Tape Library.  The
report shows all tapes registered to that User-ID.  It serves to
inform each user of the tapes currently being stored and the slot
number assigned to each tape.  This listing can be used as a
supplement to the monthly bill for foreign tape charges.

Monthly TMS Tape Report

This report provides tape volume number, data set name, data set
sequence number, volume sequence, identification of job that
created the tape, creation date, expiration date, date of last
use, and current status of the tape. The status indicates whether
the tapes are active, archived, or checked out to the user.  If
the tape was checked out of  the library, the date of removal is

To correct any entry on the  report, or to alter the status or
expiration date of any tape, the user should note the change on a
copy of the report and return it to DPSS.  (The maximum retention

NCC-IBM USER'S GUIDE                                     JAN 1988

period for an unarchived tape is 18 months.)  Any urgent
corrections to the report, including extending an expiration
date, should be made by telephoning DPSS with the appropriate
information. Users will find that effective verbal communication
regarding items on the tape report will be difficult, if not
impossible, unless the report is in hand during such

Users may request that tapes be scratched at any time by
submitting a request in writing.   The monthly tape report may be
used to note changes to be made by placing an "X" to the left of
the entry for the tape to be scratched .   When requesting that a
tape be scratched, the user should remember that all data sets on
the tape will be scratched.  If there are data sets on a tape
which must be retained, users should not scratch the tape.
Ownership will be verified prior to scratching IMS tapes.

The tape report should be returned to DPSS after all appropriate
entries, including the user's signature and commercial telephone
number, have been made.  The entries on the report must be clear
and precise.  Only in cases where there is some ambiguity will
the user be contacted by telephone for clarification or
verification of information on the tape report.

Archiving, Dearchiving and Releasing from Archive

The NCC archival library is available for offsite storage of
certain categories of users' backup tapes and tapes containing
data to be retained indefinitely, but with no immediate
processing requirement.  As of October 1, 1987, there is a $5.00
per month fee for storage in the archival library.  A user may
have a tape transferred to (archived) or returned from
(dearchived) the archive library by notifying Data Processing
Support Services (see Appendix A for telephone number).  However,
before requesting that a tape be archived, the user should first
consider archiving the data instead of the tape itself.  In this
instance the data may be archived by copying the data to disk and
request, via batch or ISO command, that the data set be archived.
Section of this manual gives detailed instructions for
disk data set archiving, or users may contact User Support (see
Appendix A for telephone number).

Requests to DPSS to archive any tape will be closely scrutinized
to ensure efficient use of NCC magnetic tape resources. DPSS will
forward all requests to archive a tape to Data Management.  Data
Management will scan the tape to ensure that 70 percent of the
tape was used.   If ample amount of the tape was used, the request
will be forwarded to DPSS for processing.

Advance notice of 24 hours is required for dearchival.  The user
will be notified when the dearchived tape is onsite and available

NCC-IBM USER'S GUIDE                                     JAN  1988

for processing.  If a job is submitted prematurely, it will be
terminated and no refund will be granted.

Archived tapes are considered to be permanently inactive, and
references to such tapes are purged from the Tape Management
System (IMS).  Inactive tapes can be read upon dearchival, but
writing to an inactive tape is not allowed.  To ensure that such
tapes are not overwritten they are protected by a file guard  ring
that prevents insertion of a write ring, and are labeled with a
"NO RING EVER" warning.  If the user wants the data on the
archived tape to become active, he should copy the data to an
active IMS-controlled library tape, and either return the
dearchived tape to the archive library or release it.  Dearchived
tapes will be returned to the archive library no later than five
days from the date of notification to the user of the
availability of the tape onsite.  If more time is needed for
processing a dearchived tape, the data should be copied to a  TMS-
controlled library tape.  Verification of ownership will be made
prior to archival, dearchival, or release from archive.

When a tape is archived, its IMS slot and associated external
label are assumed by a new active tape.   The archived tape
retains its internal VOLSER, but receives a new external label
from the archival library.  Thus, the TMS-imposed compatibility
between the internal and external VOLSERS is negated.  For
example, suppose a TMS tape with internal and external labels of
vvvvvv is archived.  The external label is changed to Dxxxxx.   In
a subsequent job that calls for the tape, the DD statement must
specify VOL=SER=vvvvw and EXPDT=98000 in the LABEL field.
However, to ensure that the proper tape is mounted, the user must
also include the following JCL tape mount statement:

            /*TAPE Dxxxxx,vvvvvv

This will display a mount message to the operator for tape Dxxxxx
and allow the system software to check the validity of the tape's
internal label against the volume serial number, vvvvvv.

It is important to remember that since the TMS handles tapes  from
the archive library like any other foreign tapes, the expiration
date in the LABEL statement should be 98000 so that TMS
processing will be bypassed.  If this is not done, the active TMS
tape with volume serial vvvvvv will be mounted instead of the
dearchived tape.

Users who have questions about tape archival or who encounter
problems with their use should contact NCC-IBM User Support (see
Appendix A for telephone number).

NCC-IBM USER'S GUIDE                                     JAN  1988

Cleaning, Evaluating, and Rehabilitating

DPSS is responsible for replacing damaged tapes, broken reels,
bands, cartridges, and label clips, as needed.  Operators are
responsible for bringing these needs to the attention of DPSS.
Upon user request, and at the convenience of the data center,
foreign tapes will be cleaned or evaluated.  The IMS provides a
list daily of scratch tapes scheduled to be cleaned/evaluated
according to volume of usage.  The usage counter for the clean
tape volume is reset at its next use.

Tape Problems

Any tape problems should be reported to NCC-IBM User Support  (see
Appendix A for telephone number) for the following reasons'-

     a.  NCC-IBM User Support will attempt to analyze the
         problem before any other actions are taken.

     b.  Problems can be isolated.

     c.  Other user reported tape problems can be compared
         and a possible trend exposed.

     d.  Evidence can be documented for possible assistance
         and refund requests.

3.4.1    TSO/E

TSO/E is a timesharing system which enables users to employ the
facilities of a computer from a low-speed terminal or from a
3270-type terminal.  The version installed at the NCC is the
basic IBM version with some enhancements such as a versatile line
editor (BED) which was developed by the Triangle Universities
Computation Center (TUCC).  (The IBM EDIT capability is available

Pertinent features of TSO/E include the following capabilities:

     a.   Text editing capabilities which allow users to
         enter, store, modify, and retrieve data at a low-
         speed terminal.   Data records can be identified by
         line number, line range, character content, or
         relative position to the last record accessed.

     b.   A remote batch facility which allows users to
         execute programs that accept input from and direct
         output to the user's terminal.

NCC-IBM USER'S GUIDE                                     JAN  1988

     c.  An extensive command procedure (CLIST) capability
         uhich allous users to build a file of executable
         TSO/E commands, subcommands, and command procedure
         statements which can be invoked as a single
         command.  CLIST's can contain symbolic variables,
         built-in functions, test and branch instructions,
         and terminal input/output.

TSO/E documentation relating to the NCC-IBM system is maintained
in the on-line data sets JUSD.TSOE.DATA, and JUSD.TSOCLIST.

TSO/E allows a user to create a JCL stream in EDIT mode and
submit it to the system for processing.  A JOB card must be
included as the first card in the stream.

TSO/E also provides the additional feature of submitting a JCL
stream stored in a data set without loading it first into a
workspace.  The MVS/Extended Architecture ISO Command Language
Reference Manual (Order No. SC28-1124) provides a detailed
description of job submission and control, or the user may issue
HELP command-name for on-line documentation.

The NCC has established performance goals for interactive
response time for ISO transactions.  These goals are=

     o   90 percent of ALL ISO transactions will be
         completed in less than 5 seconds.

     o   90 percent of SHORT TSO transactions will be
         completed in less than 1 second.

     o   90 percent of MEDIUM TSO transactions will be
         completed in less than 5 seconds.

     o   90 percent of LONG TSO transactions will be
         completed in less than 60 seconds.

Short, medium and long transactions are defined as those TSO
transactions which complete in TSO performance periods 1, 2,  and
3 or 4, respectively.  Typically, 75 percent of the Data Center's
TSO commands are short commands, 15 percent are medium commands,
and 10 percent are long commands.

NCC-IBM USER'S GUIDE                                     JAN  1988

3.M.1.1    Logon Procedures

The following interaction between the user and the system  occurs
during  the logon process of a ISO terminal session.  The user's
response is indicated by lower case characters and the system
response is indicated by upper case.  After dialing the
appropriate phone number (contact NCC Telecommunications)  the
system  will begin the session as follows:  (Also refer to  the
TYMNET  Logon Procedures in Appendix E.)

(Note;   Users in the Washington, B.C. area will not normally
encounter the dialogue on the first five lines.  They should type
at least two carriage returns to receive -the message "ENTER ISO
OR OBS" and then proceed with line six.)

P $#tt
enter  logon
logon  iii/kkkkkkkk acct(aaaaxxxxmuuu) (CR)


t - Appropriate terminal identifying character.

NNNN -  Number of the remote access node to which the terminal
session is connected.

PPP -  Number of the port of that node.

(CR)  -  Carriage return.

t## -  Number of the host computer port.

NCC-IBM USER'S GUIDE                                     JAN  1988

kkkkkkkk - Password associated with the User-ID.  The user may be
prompted to reenter the password because it has expired.  To
change it, he should enter the new password in response to the
reenter prompt.   The password may be from 4 to 8 characters long,
but may not be the same as the User-ID.  The password must be one
that has not been used the last 2 times.

acct(aaaaxxxxmuuu) - indicates account number and ADP utilization

# - Number of the news alert to reference for more details about
the message.  To reference the alert, the user may issue the
following command =

           news  alert*    ISPF

The Interactive System Productivity Facility (ISPF) is a full-
screen system application that provides users with numerous
options and panels and is executed under TSO by keying in ISPF.
Some of the options that are available are Browse, Edit,
Utilities, foreground compile and link, batch compile and link,
and TSO commands.

On-line documentation on ISPF is available under TSO HELP ISPF.
ISPF also offers the user an on-line tutorial session.
3.M . 1 . 3    $*LOGON

Your use of the IBM 3090 will be simplified by creating your own
**LOGON member.  You can create your own environment by carefully
crafting your $$LOGON member to allocate files for your needs.

Users are encouraged to create private CLIST libraries with
exactly the same Data Control Block (DCB) characteristics as
SYS2.EPA.CLIST.  Users who log on to TSO can benefit from the
following sequence of events:

     a.  The Operating System will check for a data set named
         uuuaaaa.CLIST (where uuu is your User-ID and aaaa is
         your account).

     b.  If uuuaaaa.CLIST exists, the Operating System will then
         look for a member named $$LOGON.

JAN 1988
     c.  If $*LOGON is present, the CLIST  is  executed  after  your
         logon values (User-ID, password,  account,  logon proc,
         and logon procedure) are validated and  before control  is
         returned to you.

A sample *$LOGON CLIST is as follows =

     /* LOGON PROC
      ALLOC FCSYSPROC) DSC                           +
                           £PROC                     +
                            'JUSD.CLIST1              +
                            ) SHR
      ALLOC F(SYSHELP) DSC                           +
                           £HELP                     +
                            ) SHR
     ALLOC FICISPPLIB) DSC                           +
                           SPLIB                     +
                          ) SHR
      ALLOC FCISPMLIB) DSC                           +
                           EMLIB                     +
                          ) SHR
      ALLOC FCISPSLIB) DSC                           +
                           fiSLIB                     +
                          ) SHR
     /*                      '                                      x/
     /*  CONCATENATED OBJECT.                                      */
     /*                                                            */

NCC-IBM USER'S GUIDE                                     JAN  1988

     ALLOC F(ISPTLIB) DSC                            +
                           £TLIB                     +
                          ) SHR
     ALLOC FCISPTABL) DSC                            +
                           £TABL                     +
                          ) SHR
      ALLOC FCISPLLIB) DSC                           +
                           SLLIB                     +
                          ) SHR
     /*  ALLOC FCISPLLIB) DSC                        +
     /*                       LOAD                   +
     /*                       £LLIB                  +
     /*                      ) SHR
       READ £ISPF
       ISPF £ISPF

The new user should not attempt to change this sample *$LOGON
without assistance.  An improperly structured $$LOGON can prevent
gaining access to the computer from TSO.  Of course, if you
include a LOGOFF command in your *$LOGON member, you will be
immediately logged off.

The simplest method for creating your own CLIST is to execute the
following TSO command after you have logged on:


This command will allocate a partitioned data set with the same
DCB characteristics as SYS2.EPA.CLIST.

The sample CLIST above is located in the data set
JUSD.CLISTCSLOGON) and may be copied for use in constructing a

NCC-IBM USER'S GUIDE                                     JAN  1988

Note: Under some circumstances, IBM software will create a
      PDS named CLIST (actually uuuaaaa.CLIST) with charac-
      teristics of variable length, blocked, and logical
      record length (LRECL) of 255.  This type of structure
      wastes all the bytes after column 80 and is incom-
      patible with the structure of the system CLIST
      data sets at the NCC.  It will generate a SYNAD error
      when a CLIST from such a CLIST partitioned data set
      is executed.

      Instead a data set with characteristics of fixed block
      and a logical record length of 80 bytes is most
3.4.2    CICS

Customer Information Control System (CICS) as implemented at the
NCC is a general purpose data base/data communications interface
between the operating system and application programs written in
VS COBOL II or PL/1.  CICS performs control functions necessary
to manage a network of local and remote terminals and subsystems
and to manage a data base (ADABAS). which may be shared with other
users in other partitions or regions of the operating system.

CICS is a controlled, shared resource.  A large number of users
accessing a variety of application programs will operate in a
single MVS environment under the control of the central data base
administrator.  The current interactive service level objective
for CICS is that 90 percent of all CICS transactions should be
completed in less than 2 seconds.

Documentation in the form of the CICS Application Development
Procedures Manual is available from the CICS System Administrator
at the telephone number in Appendix A listed under NCC User
Support.  For further information contact the CICS System
Administrator or the Central Data Base Administrator (telephone
numbers are listed in Appendix A).

NCC-IBH USER'S GUIDE                                     JAN  1988


Remote Job Entry (RJE) access to NCC-IBM is available utilizing
both dial and dedicated phone circuits.  The NCC-IBM supports
HASP as well as SNA RJE's.  Users should contact the NCC Network
Support Group (see Appendix A for address and telephone number)
to discuss access and to request registration forms, as well  as
to get information on the location of any current RJE site.   The
user should complete the form, using the directions which are
included, and return it to NCC.  When NCC receives the completed
form, the RJE will be configured and a confirmation memorandum
indicating the remote number and associated password will be
mailed to the user.

For sites with infrequent access needs, a method of pooling
remotes is provided which permits the sharing of remote ID'S
among sites.   This pooling lessens resource requirements and
minimally affects the user.  Any questions concerning RJE usage
at NCC-IBM should be directed to the NCC Network Support Group
(see Appendix A for telephone number).


JES2 provides an interface for jobs to be entered and processed
by an IBM MVS system.  These jobs can be entered through a remote
batch work station or TSO.  These are some of the functions
performed by JES2 :

     o   Reading jobs and SYSIN data, both local and remote

     o   Spooling jobs and input data to direct access

     o   Reading SYSIN data and writing SYSOUT data for
         active jobs

     o   Writing jobs and SYSOUT data, both local and remote

     o   Storing jobs on an internal job queue

     o   Controlling access and operations for RJE stations

Appendix C provides a list of JES2 Control Statements.   Appendix
D contains a list of JES2 operator commands.

NCC-IBM USER'S GUIDE                                     JAN  1988



M.1.1     JOB Statement

The major unit of work performed on the NCC-IBM computer is
called a job.  Jobs consist of one or more steps, each of which
executes a procedure or a program.  The purpose of the job and
step  arrangement is to force programs which relate to each other
to  run in a predefined sequence.  As an example, a job may
consist of:

     o   A step that creates a data set.

     o   A step that sorts that data set.

     o   A step that reports on the contents of the sorted
         data set.

Since  the order of step executions is controlled, status
information may be passed between steps allowing steps to be
executed conditionally, depending on the success or failure of a
single preceding program or a series of programs.  Data sets may
also  be passed forward to successive steps of the job.

Users  should be aware that every OS job is required to have a
valid  JCL JOB statement as its first control statement or all of
the JCL will be ignored.  The JOB statement provides the
operating system with basic data such as =

     o   Billing information.

     o   Priority for which the user is willing to pay.

     o   Amount of main storage required by the job.

     o   Time limitation after which the job will terminate.

The following is a description of the NCC-IBM standard JOB card
format, a discussion of the individual parameters and their
default values (if any), and a description of the way in which
the NCC deals with priorities and the queuing of jobs.

The following sample JOB statement illustrates the format of the
NCC-IBH JOB card.  Required fields are User-ID, account with UID,
room,  programmer, and password.  Passwords are also required for
jobs  submitted via RJE.

NCC-IBH USER'S GUIDE                                     JAN  1988

    //iiiaa JOB (acctxxxxmuuu,room,etime,lines,cards,forms , copies ,
    //          log,linect),programmer,MSGCLASS = s ,
    //          MSGLEVEL=(n,n),TIME=(m,s),PRTY=p,REGION = xK ,
    //          COND=(c,op),TYPRUN=tttt,NOTIFY=iii,RD=d,
    //          PASSWORD=xxxx

iii should be replaced with the user's three-character NCC-IBM

The IBM publication MVS/Extended Architecture  JCL Reference
(Order No.  GC28-1352) contains a detailed explanation of  the JOB
card fields and subfields.

The fields are described:

aa          This subfield,  combined with the User-ID creates  a
            unique MVS job name.  It will be assigned
            automatically by the system if it  is  not supplied.

acct        Required positional parameter.  Consists of a  valid
            account code (acct) with a valid ADP  Utilisation
            Identifier (xxxxmuuu).

xxxxmuuu    The ADP Utilization Identifier used to identify a
            specific ADP system or activity and to associate
            computer usage statistics with that activity.  xxxx
            is an acronym identifying the system; m is the mode
            character which may be one of the  following:

            F - Computer runs are to study the feasibility of
            the activity.

            D - Computer runs are to develop software to
            support or implement the activity.

            T - Computer runs are to test the  software associated
            with the activity.

            P - Computer runs are to use the software associated
            with the activity.

            M - Computer runs are to maintain  the software
            associated with the activity.

            R - Computer runs are to retrieve  data associated
            with the activity.

            U - Computer runs are to update data  associated
            with the activity.

            uuu are optional characters which  may be assigned


                                             JAN  1988
            by a user to further subdivide or classify computer
            usage.  If specified, these characters may be

room        Positional parameter tells DPSS and WIG Production
            Services where to file or mail printed output.
            Room codes are explained in Section M.S. 6. 2.

All of the optional subfields referenced ("etime"-lllinect") need
only be supplied when the user desires to override the
established default.  All leading commas must be present since
the subfields are positional.

The JES2 '/*JOBPARM' card, inserted between the JOB card and the
first EXEC card, may be used to specify the positional subfields
Iletime"-"linect11 .   The required operands are specified as  keyword
parameters.  See Appendix C for the proper statement format.
Optional positional parameter.  Specifies, in minutes
the amount of elapsed wall-clock time that the job
may use.  The format is one to four digits.  If the
expressed time is exceeded, the system prints a
warning message, but the job continues to run.  This
is not CPU time.  Maximum is 1,440 minutes.

Optional positional parameter. Indicates the maximum
number of lines to be printed by the job, expressed
in thousands.  The default is 20 (20,000 lines).  The
maximum is 9999 (9,999,000 lines).

Optional positional parameter.  Up to four digits;
indicates the number of cards to be punched in
hundreds.  The default is 20 (2,000) cards.  The
maximum is 999,900.

Optional positional parameter. Indicates the type of
form desired for printed output.  See Section 4.3.8
for details about forms.  If not specified, Form 6111
will be used.
Optional positional parameter.  Indicates the number
of printed output originals desired.  The default is
1 .
Optional positional parameter.  Governs production of
the JES2 log.  Expressed as N for no log and Y for
printing of the log.  The default is Y.

Optional positional parameter.  Specifies the number
of lines to be printed on each page of printed
output.  The default is 61.

                                                         JAN  1988
            Required positional parameter.  Contains the
            programmer's name or initials.  If this field
            contains special characters, such as blanks, periods
            or hyphens, the string must be enclosed in quotes.
            The maximum length is 20 characters.

            Required parameter on JOB card of jobs submitted from
            an RJE station, or where the User-ID of the
            submitting user is not an exact match to the User-ID
            of the submitted job.

The following are commonly used JOB card keyword parameters which
NCC users may find useful.  All are optional with no positional
requirements =
                 Specifies the output class to which system
                 messages and JCL statements for the job are to
                 be written.   The' default is MSGCLASS=A.

                 Indicates what job output is to be written as
                 part of the output listing.  The default is
                 MSGLEVEL=(1,1),  which provides all JCL
                 statements and allocation/termination messages.

                 Specifies the maximum amount of CPU time in
                 minutes (m)  and seconds (s).  The default is 7
                 seconds (TIME=(,7)). When possible, TIME should
                 be specified in seconds.

                 Specifies the priority for which the user wishes
                 to be billed; default is 2.  See Section 3.3.2.

                 Specifies the maximum amount of main storage
                 space to be allocated to a job= e.g. REGION=10K
                 will request 10,000 bytes of storage.  If not
                 specified, the default is 1000K.  (See items
                 "K", "Megabyte" in Glossary.)

                 Controls job processing based on return codes
                 issued by one or more job steps.

                 Indicates special JES2 processing; e.g.,
                 TYPRUN=SCAN indicates that the JCL is to be
                 scanned for syntax errors but the job will not
                 be executed.  TYPRUN=COPY indicates that the
                 input deck as submitted is converted directly to
                 a SYSOUT data set and scheduled for output
                 processing.   The class of the SYSOUT data set is
                 the same as  the MSGCLASS of the job.

                                                JAN 1988
RD =
        Used to request a message be sent to a user's
        TSO terminal when a batch job has completed

        Specifies now the step restart facilities are
        used and whether automatic restart is permitted
        or suppressed. See MVS/XA JCL manual (Order No.
        GC28-1352) and MVS/XA Checkpoint/Restart for
        information on the RD parameter.
M. 1 . 2
Execute (EXEC) Statement
The EXEC statement is the first statement of every job step and
procedure step.  It identifies the program to be executed or the
catalogued procedure to be called.

The format is =

        //stepname EXEC PGM=progname,PARM='pp...p',RD=d,REGION=xK,
        //              TIME=(m,s),COND=(c,op,stepname),
        //              ACCT=acctxxxxmuuu

The EXEC statement can also be used to assign a step name.  The
step name identifies a job step within a job.  The step name is
coded in the name field of the EXEC statement (starting in column
three), and must consist of one through eight alphanumeric and
national characters (
NCC-IBM USER'S GUIDE    Parameters
JAN 1988
There are several useful parameters which may be specified as
operands on an EXEC statement.  These parameters and their
functions are discussed belou.

FARM     Hay be used to pass up to 100 characters of information
         to the processing program at execution time.

RD       Specifies how the step restart facilities are used with
         the checkpoint macro and whether automatic restart is
         permitted or suppressed.  Refer to the IBM JCL Reference
         Manual for more details.

REGION   Used to specify how much main storage, in bytes, is to
         be allocated to the job step. If this parameter is not
         coded on either the EXEC or JOB statement, the system
         default region size (1000K bytes) is used.  If the
         REGION parameter is coded on the JOB statement, all
         REGION parameters coded on the job's EXEC statements are
         ignored.  If the REGION parameter is coded on an EXEC
         statement that calls a procedure (PROC), the REGION
         parameter on the calling EXEC statement is used.  The
         REGION parameters within the PROC steps are ignored.

TIME     Can be used to specify the maximum amount of time the
         step can use the CPU.  If not specified, the maximum
         time allowed in a step is 60 minutes.  If more than 60
         minutes of CPU time is needed for a step, the TIME
         parameter must be used in the step, as well as in the
         JOB card.  The time specified on the JOB card indicates
         the maximum time allowed for all steps in the job.  By
         coding the TIME parameter, the amount of CPU time wasted
         can be limited.

COND     Can be used to eliminate unnecessary use of computing
         time by basing the execution of a job step on the
         successful completion of one or more preceding steps.
         This parameter can be used to specify either conditions
         for bypassing a job step or conditions for executing a
         job step even if preceding job steps have abnormally
         terminated.  If this parameter is also specified on the
         JOB statement, any conditions tested on the JOB
         statement take precedence over those specified on the
         EXEC statement.

ACCT     Can be used to respecify the account code (acct) with or
         without the ADP Utilisation Identifier (xxxxmuuu) for
         the job step.  If this parameter is not coded on the
         EXEC statement, the account code and the ADP Utilization
         Identifier coded on the JOB statement will be associated

NCC-IBM USER'S GUIDE                                     JAN  1988

         with the job step.  To respecify only the account code,
         only the acct characters need be coded.  However, if the
         ADP Utilization Identifier is to be respecified, both
         the account code and the ADP Utilization Identifier must
         be coded.   Refer to Section 4.1.1 for detailed
         instructions for coding the ADP Utilization Identifier.

Each of the above parameters may also be specified on an EXEC
statement calling a catalogued procedure. Refer to Section 4.2.3
for a discussion of the effects of coding parameters on an EXEC
statement invoking a catalogued procedure.

Many other EXEC-statement parameters may be coded if required,
but their description is beyond the scope of this manual.
Complete instructions for coding EXEC-statement parameters are
given in the IBM MVS Extended Architecture JCL Reference Manual
(Order No. GC28-1352).    JOBLIB and STEPLIB Considerations

The system normally searches a standard set of system libraries
for executable programs specified on EXEC statements in the Job
Control Language.  If a program to be executed resides in a
private library,  a JOBLIB or STEPLIB DD statement must be
supplied giving the appropriate data set name.

If a JOBLIB statement is used, it applies to every step in the
job.  Each time a program is requested, the system looks first in
the library specified on the JOBLIB DD statement; if it does not
find the program there, it then searches the system library.  The
JOBLIB statement must precede all EXEC cards.

A STEPLIB DD statement applies only to the step in which it is
included.  For that step, the system will look first in the
specified library; if it does not find the program there, it will
then look in the system library.  The STEPLIB statement must
immediately follow the EXEC card for which it applies.

If both JOBLIB and STEPLIB DD statements appear in a job, the
STEPLIB definition has precedence.  The library defined by the
JOBLIB DD statement is not searched for any step that contains a
STEPLIB DD statement.

Since a JOBLIB DD statement causes the private library to be
searched during each job step, regardless of whether the desired
program resides on the system library or a private library, it is
frequently desirable instead to specify a STEPLIB DD statement
for each step where a private library is needed.  This eliminates
the expense of searching a private library for a program that
resides in the system library.  Furthermore, STEPLIB DD

NCC-IBM USER'S GUIDE                                     JAN  1988

statements offer the added advantage of allowing the runstreams
in which they are contained to be concatenated and submitted  as
one job.  This is not possible with two runstreams containing
JOBLIB DD statements, as an input jobstream may only contain  one
JOBLIB DD statement.
4.1.3    Data Definition (DD) Statement

The DD (data definition) statement describes a data set that is
to be used in a job step and specifies the input and output
facilities required for use of the data set.  Each data set that
is to be used in a job step requires a DD statement; all DD
statements for a step follow that step's EXEC statement.  Each
job step may include no more than 1635 DD statements.

The DD statement consists of the characters "//"', and four
fields, the DD name, "DD", the required operands, followed by
comments, if desired.  The rules for coding the DD statement are

     a.  The characters "//" in positions 1 and 2.

     b.  The DD name (which must be unique within the job
         step, must consist of 1 through 8 alphanumeric and
         national (<3 tt $) characters, and must not begin
         with a numeric character).   The DD name must start
         in column 3.  No DD name can be used if
         concatenation to the preceding liii statement is

     c.  At least one blank follows the DD name (or "//"' if
         a DD name was not coded).

     d.  "DD".

     e.  At least one blank follows DD.

     f.  An operand (positional parameter or keyword

     g.  All other required operands (keyword parameters),
         separated with single commas.

     h.  At least one blank.

     i.  Any desired comments.

NCC-IBM USER'S GUIDE                                     JAN  1988

     j.  To continue a DD statement onto the next line, the
         user may end any operand with a comma in or before
         Column 71, start the next line with a // and at
         least one blank, and continue listing necessary
         operands.  The first of these must start on or
         before Column 16.
In addition to the DD name coding rules given in b. above,
several DD names have special meanings.  These DD names and their
meanings are discussed in detail in the IBM MVS Extended
Architecture JCL Manual (Order No. GC28-1352) in the section
entitled DD Statement.

There are two categories of operands which may be specified on DD
statements:  positional parameters, which must precede all other
operands, and keyword parameters, which may be coded in any
order.  The only positional parameters are '*', 'DATA1, 'DUMMY1,
or 'DYNAM*.  Lists of both types of DD statement parameters, and
a discussion of their use, may be found in the IBM MVS Extended
Architecture JCL Manual (Order No. GC28-1352).

When using card data as input which has // in columns  1 and 2,
the DD card must be as follows:


If the card data has /* in columns 1 and 2, a delimiter parameter
should be coded as follows:

       (card data)    Disk Data Sets

The simplest form  of a DD statement describes a data set that
already exists:

    //SYSUT1  DD  DSNAME=iiiaaaa.name1,Di~P=SHR

For various reasons, there are occasions when users create data
sets which remain empty.  NCC policy states that any data set
remaining empty for seven days after its creation or last access
will be deleted.  Users are encouraged not to create data sets
ahead of the time they are needed.  It is best to create them new
in the JCL.

On the other hand, a DD statement describing a new data set
requires significantly more detail:

NCC-IBM USER'S GUIDE                                     JAN  1988

    //SYSUT2  DD  DSN = iiiaaaa.name2,DISP=(NEW,CATLG,DELETE) ,
    // UNIT = DISK,VOL = SER=USR005,SPACE=(6160, (60, 18) ,RLSE),

Descriptions of these statements  and the most commonly used
parameters and subparameters follow:

The DD name is SYSUT1 in the first example; SYSUT2 in the second.
These DD names are commonly used  in IBM utilities such as
IEBGENER for input and output data sets, respectively.

DSNAME (or DSN) - Specification of the data set name.  If this is
not specified, the system will assign a name.  This is
appropriate for new data sets to  be used only throughout the
duration of the job; that is, temporary data sets.  When the name
is specified, it must be the fully qualified data set name (see
Section 3.3.1) not a ISO-type name format.

DISP - The first subparameter describes the current status of the
data set (old, new or to be modified); the second and third
subparameters specify the disposition of the data set upon normal
and abnormal termination, respectively, of the job (pass, keep,
catalog, uncatalog, or delete).  If the third subparameter is not
given, it defaults to the second.   The various options are=

                     NEW   KEEP      KEEP
                     OLD   DELETE    DELETE
              DISP= (SHR ,  CATLG  ,   CATLG)
                     MOD   UNCATLG   UNCATLG

If no first subparameter is specified, NEW is assumed.  If DISP
is not given at all, DISP=(NEW,DELETE ) is assumed.  DISP=(,CATLG )
is equivalent to DISP=(NEW,CATLG).  All uncatalogued and non-
standard named data sets are regularly deleted.  The user should
use DELETE as the third (or abnormal termination) subparameter of
the DISP parameter when creating  new data sets .  This precaution
avoids leaving behind unwanted data sets if a job abnormally
ends .

DISP=OLD and DISP=SHR both  imply  existing data sets.  The user
should exercise caution in  indicating OLD since this gives
exclusive control to this data set.  That is, no other job can
access that data set through the  duration of the job step nor
will the step start execution while any other job or ISO session
has access to that data set.

NCC-IBM USER'S GUIDE                                     JAN 1988

With the shared DASD software package being used on the IBM-NCC
system, TSO users may receive additional messages which will
inform them of any exclusive control conflicts.  These messages
will begin with MSI501I or MSI503I and will indicate the job
name, system task name, and/or sessions which have conflicts.

SHR permits existing data sets to be accessed concurrently.  The
second subparameter defaults to KEEP for either.  Use of /*CNTL
may also be used to prevent contention (see Appendix C).

DISP=MOD is applicable only to sequential data sets.   It may be
specified for either new or existing data sets.  If the data set
exists, it will be opened for processing just after the last
record in the data set, thus providing a means of adding new
data.  The second subparameter defaults to KEEP for DISP=MOD.

DISP=(,PASS) is used to allow a data set to be passed to
subsequent job steps.  Advantages of passing a data set are:

     a.  The system retains data set location and volume
         information across step boundaries.

     b.  Subsequent job steps can refer back to the data set
         by name.

DISP=(,KEEP,DELETE) will retain the data set on the volume but
will not make an entry in the catalog.  DISP=(,CATLG,DELETE)
should be used for any USRxxx data set at NCC.

UNIT - Specification of the particular I/O device to be used.
(See Section for different device specifications
available at NCC).  UNIT=DISK is recommended for any direct
access disk storage device available at NCC.

VOL=SER - This should not be specified when space is desired for
a catalogued data set on a USRxxx pack or a temporary data set on
a public work pack.  The system will select an appropriate disk
volume depending on the space requirements.  In the second
example above, it would be better to leave it out.  Given the
specific volume reference shown, only volume USR005 would be
investigated.  If sufficient space is not available,  the job will
abnormally terminate.

If UNIT or VOLUME is specified, the system bypasses the catalog
in trying to determine a data set's location.  If the volume is
specified, the unit must be specified for an existing data set.
Also, if the unit is specified, the volume must be specified.
Therefore, if the user is referring to an existing catalogued
data set, he should only code the DSNAME and DISP parameters.
                              4-1 1

NCC-IBM USER'S GUIDE                                     JAN 1988

SPACE - This has a number of considerations and consequently is
addressed in significant detail in Section following.
In the example given, SPACE=(6160,(60,18),RLSE), 60 blocks of
6160 bytes each will be allocated.   The RLSE parameter should
always be used on new data sets thus releasing unused space at
the end of the job.

DCS - There are many possible subparameters to describe the data
control block (DCB).  Only the most common are described below.

RECFM - specifies the record format.  The first character may be
any one of the following:

     U - Undefined record format.  LRECL should not be
     coded. BLKSIZE must be greater than or equal to the
     largest block written.  When reading from or writing
     to this data set, entire blocks are referenced.  They
     may be any size up to the BLKSIZE.  It may only be
     used in conjunction with T, A, and M (see below).

     V - Variable length records.  LRECL must be the length
     of the largest logical record plus four bytes for a Record
     Descriptor Word.  BLKSIZE must be at least as large
     as LRECL plus four bytes for a Block Descriptor Word.

     F - Fixed block records.  BLKSIZE must be a multiple
     of LRECL or equal to LRECL for unblocked records .

     D - Variable length ASCII tape records.

The following characters may be used in any combination after the
first character:

     B - Blocked records (only used with F or V in the first
     character).  Blocking forms a physical block made up of
     a number of logical records.  An inter-record gap is
     recorded between each block written; therefore, a block is
     recognizable by the hardware device.

     T - Track overflow.  A track is completely filled even
     though a block must be split across tracks.  Used when
     the block size exceeds the track size.  CONTIG must
     be used in the SPACE allocation (see Section
     Cannot be used with D.

     S - When used with F (FBS or FS) this indicates fixed,
     standard format.  This is acceptable only for sequential
     data sets and indicates that there are no truncated blocks
     (blocks less than BLKSIZE) with the possible exception of
     the last.  For VS or VBS this indicates spanned records and
     implies that logical records can be spanned across block

NCC-IBM USER'S GUIDE                                     JAN  1988

     boundaries.   Cannot be used with U, T, or D. DISP=MOD should
     not be used with FS or FBS.

     A - ASA carriage control characters are in the first
     byte of each record.  Used to control the printer.
     Cannot be used with M.

     J1 - Machine carriage control characters are in the
     first byte of each record .   Also used to control the
     printer.  Cannot be used with A.

LRECL - This is used to specify the length of the logical record
for fixed or variable-length records.

BLKSIZE - The specification of the size of each physical record
in bytes.   The maximum blocksize for a disk data set is one
track. The 8380 disks at NCC have a track size of 47,476 bytes.
However, due to various software limitations, only programs
issuing their own EXCP's directly may obtain block sizes above
32760.    Allocating Direct Access Storage Device (DASD) Space

All new data sets on direct access storage devices must have
space allocated.  To make the most efficient use of disk space,
users must know something of the devices available to them,
including the relationship of data blocks to the tracks and
cylinders on these devices.

There are three basic methods of allocating space; by cylinder,
track, or bytes per block. For the TSO format used to allocate
data set space, the user may issue "HELP ALLOCATE" from a TSO
session. In BATCH, the SPACE parameter is used in the following

           CYL                          index
SPACE=(     TRK   ,(primary,secondary,directory))
     Primary - The number of units (blocks, tracks, or
     cylinders) to allocate.  It may be allocated in
     up to five extents (separate physical locations
     on the disk) and each will always be on one volume.

     Secondary (optional) - The number of units to
     allocate when the primary is full.  Both the primary
     and secondary must be large enough to contain the largest
     block written.  There may be a maximum of 16 extents

NCC-IBM USER'S GUIDE                                     JAN  1988

     allocated on one volume for a data set.  This includes
     all primary and secondary extents.  If used, the secondary
     extent allocation should be smaller than the primary.

     Index - Reserves space for an ISAM data set.

     Directory - Reserves space for the member names in
     a partitioned data set.  This space will be taken
     from the primary allocation.  If directory space is
     given, the system assumes DSORG=PO; if not given,
     DSORG=PS is assumed.

The primary space is allocated in its entirety when the data set
is created. The secondary is allocated only after the primary is
exceeded.  Allocating with a secondary offers the following

     o   The primary amount can be conservative since the
         secondary provides a reserve.

     o   The secondary provides growth potential over time.

A secondary space allocation may be specified in a DD statement
for an existing data set. This could be used in the following

     o   No secondary was specified when _ne data set was

     o   The original secondary allocation is to be

In both cases the new secondary increment is valid only through
that step and does not change the original allocation or the
limit of 16 extents.

Because of this ability to override original space allocations,
care must be taken when a DD statement containing a SPACE
parameter is to be overridden. The user should code "SPACE=" on
the overriding statement to nullify the SPACE parameter.

Users should allocate only the space required for the current
data set size and use a secondary allocation if expansion is
expected.  Including the RLSE subparameter (see below) in the
SPACE parameter for sequential data sets releases unused space on
new data sets although the secondary allocation is still
available if needed.  Overallocations in current data sets can be
released by adding the RLSE subparameter to any job opening the
data set or by the $SH command in ISO (see HELP $SM under TSO for

NCC-IBM USER'S GUIDE                                     JAN 1988

Allocation by Blocks

Allocating by blocks is the closest to device-independent
allocation that is available. The system will compute the number
of tracks needed based on the number of blocks that completely
fit on a track. For optimum use of space, the user still must be
aware of device characteristics (see Finding Optimum Blocksize in
this section).

The bytes per block indicated in the SPACE parameter should equal
the DCB=BLKSIZE subparameter so that the system can include
inter-record gaps in its calculation.  When secondary space is
allocated, the system uses the DCB BLKSIZE value rather than the
blocksize given in the SPACE parameter.  If blocks have keys, the
KEYLEN should also be coded in the DCB for correct space

Allocation by Tracks or Cylinders

The TRK or CYL specifications are similar to the blocksize
specification in that they define the unit of measure to be used
when disk space (an extent) is allocated.  TRK says that each
extent allocated will be 47476* (primary or secondary) bytes
long.  CYL says that each extent allocated will be 15*47476*
(primary or secondary) bytes long.

Allocating Directory Space for Partitioned Data Sets (PDS)

The directory is always allocated in units of 256-byte blocks, no
matter what unit of allocation is specified in the SPACE
parameter.  A rule of thumb in determining the number of
directory blocks to assign is to estimate 5 members per directory
block. For example, if the user anticipates 52 members total, he
may divide 52 by 5 and round up to the nearest integer -- 11
directory blocks.  The user should allow sufficient growth  room
in this allocation. No secondary is given for a PDS directory; if
the user runs out of directory space, the data set must be
recreated with a new space allocation.  The user may copy the old
data set into a newly allocated one, or he may use the IEHMOVE
utility to expand the directory space.  L^ace for the directory
is allocated in units of tracks and any unused space on the last
directory track is wasted unless there is enough room for a block
of the first member.

Space Subparameters RLSE, CONTIG, and ROUND

Several SPACE subparameters are available to release unused space
and to allocate space for faster access.  All of the
subparameters are positional; therefore, commas are needed  to
account for their absence.

NCC-IBM USER'S GUIDE                                     JAN 1988

        CYL          ind

RLSE releases all unused space when the data set is closed.  This
frees the user from the need to compute required space since any
wasted space is freed at the end of the job step.  If an extent
is allocated and only partially used, the remainder of the extent
is released. If space was allocated in cylinders, only unused
cylinders are released; allocation by blocks or tracks releases
unused tracks.  Space can be released for existing data sets by
receding the unit of measurement and primary quantity exactly as
they appeared in the original request.  For example, if the
original request was "SPACE=(TRK,(100,50))" unused tracks can be
released when the data set is retrieved by
"SPACE=(TRK,(100),RISE)."  Remember that the data set must be
opened and closed before space is  released; i.e., invoking
IEFBR1U will not cause the release of space.

Some caution must be observed when using the RLSE parameter for
partitioned data sets.  Every time a member of a partitioned data
set is replaced, additional space  is used.  The space occupied by
the old member becomes inaccessible until the partitioned data
set is compressed.   If a PDS is not compressed periodically to
release this space, a considerable amount of space is wasted.  A
PDS may be compressed with IEBCOPY (see Section 5.3.5) or with
the $SM command in ISO or batch.

CONTIG is used to allocate contiguous space and should be  used
for high-access data to minimize access time.  If track overflow
is used, CONTIG should always be coded.  Contiguous space
allocates all of the primary space in one extent.  Secondary
space may not be contiguous.

ROUND is used to allocate space on cylinder boundaries when space
is requested by blocksize.  The system computes the space needed,
rounds up to the nearest cylinder, and allocates complete
cylinders, starting and ending on a cylinder boundary.  This
allocates the minimum number of cylinders necessary and decreases
access time to the data.

NCC-IBM USER'S GUIDE                                     JAN  1988

Finding Optimum Blocksize

The main reason that a user is allowed to specify blocksizes  for
data sets is that only he can decide which aspect of data set I/O
is more important, disk space utilization or program core
utilization.  In general, the larger the blocksize, the less
wasted space there is on a data set and fewer I/0's are required
to access the data set. Smaller blocksizes require less core  in
the application program, sometimes significantly less since MVS
usually uses five buffers of one block each for each data set.
Figure 4-1 shows these effects for a data set with 10,000 80-byte

    Blocksize:            80        8000
    Recs/Block:            1         100
    Blocks/Track:         83           5
    Recs/Track:           83         500
    Tracks in DS:        121          20
    Buffer Space:        MOO       40000
    #I/0's:             10000         100

        Figure 4-1.   Data Set with 10,000 80-Byte Records

Figure 4-1 uses a blocking factor of 100 that is commonly used
because it is convenient.  A proper blocksize optimization should
consider some of the physical aspects of the 8380 disk drives.
The user should note that the 8380's write data in 32-byte
increments.   An 80-byte block will require 96 bytes on disk (3
increments).  Twelve bytes are used by the disk for error
checking for each physical block's data and key areas (data only
if KEYLEN=0).  The following formulas should be helpful in
blocksize calculations:

                 ELSE OVERHEAD--=22

                 ELSE KEYSPACE:=(KEYLEN+12)/32


The results of the divisions should be rounded to the next
highest integer.

OVERHEAD is the number of 32-byte increments required for
physical disk record overhead requirements.

KEYSPACE is the number of 32-byte increments required for key

NCC-IBM USER'S GUIDE                                     JAN  1988

DATASPACE is the number of 32-byte increments required for data

The number of blocks that can be stored in one track can be
computed with the following formula:


In this result,  the fractional portion should be truncated.   The
number of 32-byte increments in one track is 1499.

The number of records per track can be computed with the
following formula:

     NUMREC:=NUMBLKS*(number of records per block)

By computing the number of records per track for various
blocksizes, the user can determine the optimal blocksize for  a
data set by balancing the efficiency of disk space utilization
against core requirements.  Using the example in Figure 4-1,  the
most efficient use of disk space would occur with a blocksize of
23440 (or 293 records per block) which would yield 586 records
per track.  The core cost to the program would be five
23,440-byte buffers.  By reducing the blocksize to 11440 (or  143
records per block) the core buffer requirements are reduced by
more than one-half and there are 572 records per track.  Both
data sets would require 18 tracks to store the sample  10,000

Calculations of this sort are ideally suited to a personal
computer and a 10-line Basic program.  If the user does not have
access to these facilities, Figure 4-2 provides various maximum
blocksizes for the given number of blocks per track.

                                              JAN 1988
1 1
W/0 Keys
1 1 ,476
3, 188
2, 932
1 ,876
1 , 396
1 ,076
23, 240
1 1 ,240
6, 12C
4, 040
1 ,640
*The specifications for 1 block per track
are omitted due
which prohibit
32,760 bytes.
to software
a blocksize

greater than

               Figure 4-2.   Determining Block Size

These and other subjects related to JCL are covered extremely
well in the book  "SYSTEM/370 JOB CONTROL LANGUAGE".  It is by
Gary DeWard Brown and is published by;

                   John Wiley £ Sons
                   605 Third Avenue
                   New York, NY 10016
Tape Data Sets
Magnetic tape capabilities are provided for the off-line storage
of extremely large and infrequently used data sets.  A full
2400-foot reel of tape recorded at a density of 6250 bits per
inch can record more than 160 million characters when blocked at
the maximum of 32,760 bytes per block.

NCC-IBM USER'S GUIDE                                     JAN  1988

Magnetic tape data sets may be catalogued just as disk data sets;
i.e., a DD card defining a catalogued tape data set need only
specify data set name and disposition:

   //anyname DD DSN=iiiaaaa.tapequal,DISP=OLD

The following example creates a tape data set, also referred  to
as a tape file =

  //anyname DD DSN = iiiaaaa.tapequal,DISP=(NEW,CATLG,DELETE ),
  // LABEL=(1,SL,,OUT,EXPDT=84240)

Descriptions of these statements and the most commonly-used
parameters and subparameters follow;

The DD name (anyname) may be any valid name as specified in
Section 4.1.3.  If an IBM utility such as IEBGENER is used, the
DD name will be SYSUT2 denoting the output data set.

DSNAME (or DSN) - Specification of the data set name.  If  this is
not specified, the system will assign a name.  This is
appropriate for new data sets to be used only throughout the
duration of the job; that is, temporary data  sets.  When the  name
is specified, it must be the fully qualified  data set name  (see
Section 3.3.1) not a TSO-type name formaL.  If a tape data  set is
catalogued in the system catalog, normal NCC  naming conventions
must be maintained (see Section 3.3.1).  However, if the data set
is not to be catalogued, any name conforming  to IBM standards
(see MVS/XA JCL reference manual) may be used.  The Tape
Management System will save the entire data set name in the Tape
Management Catalog, whether or not the name is catalogued  in  the
System Catalog.  Therefore, a data set name should be assigned to
every data set that the user intends to save.  If no name  is
specified when the data set is created, the temporary data  set
name is saved in the TMC.  The next time the  user wants to  use
the data set he must look up the temporary name (either in  the
creating job or by searching the TMC) and use that name.   This is
also true for unlabeled tapes.

DISP - The first subparameter describes the current status  of the
data set (old, new or to be modified); the second and third
subparameters specify the disposition of the  data set upon  normal
or abnormal termination, respectively, of the job (pass, keep,
catalog, uncatalog, or delete).  If the third subparameter  is not
given, it defaults to the second.  The various options are=

NCC-IBM USER'S GUIDE                                     JAN  1988




If no first subparameter is specified, NEW is assumed.  If DISP
is not given at all, DISP=(NEW,DELETE) is assumed.  DISP=(,CATLG)
is equivalent to DISP=(NEW,CATLG).

Since tapes cannot be used concurrently between jobs as disk data
sets are, DISP=OLD giving exclusive use of a tape data set to a
job does not usually present a danger.  However, the user should
remember that the exclusive use is on the data set name.
Therefore, if a data set on tape has a name of SYS1.HELP, no
exclusive use is permitted since the system data set SYS1.HELP
which resides on disk is always allocated by TSO.

DISP=MOD may be specified for either new or existing data sets.
If the data set exists,  it will be opened to point after the last
record already written providing a means of adding new data.  The
second subparameter defaults to KEEP for DISP=MOD.

DISP=(,PASS) is used to allow a data set to be passed to
subsequent job steps.  Advantages of passing a data set are=

     a.  The system retains data set location and volume
         information across step boundaries.

     b.  A mountable volume containing the data set remains

     c.  Subsequent job steps can refer to the data set by
         name .

     d.  The file number and label type of a tape data set
         are passed.  (Density is not.)

DISP=(,KEEP) will retain the data set on :he volume but will not
make an entry in the catalog.  DISP=(,CATLG) will make an entry
in the catalog for the tape data set with volume serial, file
number, and unit specification.

UNIT - Specification of the particular I/O device type to be used
for different device specifications at NCC.   DEFER specifies that
the system should allocate a device for the data set, but the
volume(s) for that data set will not be mounted until the data
set is opened.   The advantage of this is that, in the event that
a program terminates before a tape is needed, no tape volume will
be assigned since no tape mount has been issued.

NCC-IBM USER'S GUIDE                                     JAN 1988

The following chart shows the permissible unit types for 7- and
9-track tapes.  The type of tape drive requested will depend on
the tape itself (7- or 9-track) and the density desired.  Density
will default to 4 (6250) for all tapes on 1600/6250 drives.
Permissible unit types are depicted in Figure 4-3 below:

TRACK         DENSITY            DEVICE NAME

7-track       200/556/800 bpi    TAPE7, 2400-2
9-track      *800 bpi/1600 bpi   TAPES, 2400-4
9-track       1600/6250 bpi      TAPE, TAPE16, TAPE9, TAPE62,
                                 2400, 3400-5

*800 bpi specifications can only be obtained with JCL giving

                Figure 4-3.  7- and 9-Track Tapes

Occasionally a problem results from the two types of 9-track tape
drives:  one handling densities of 800 and 1600; the other, only
1600 and 6250 bpi.  No problem will occur if users indicate those
unit types  defined by NCC:

     TAPES  for 800/1600

     TAPE/TAPE16/TAPE62 for 1600/6250

If the IBM  generic unit types are specified (e.g., 2400, 2400-3,
or 3400-3)  problems could occur if the following conditions

     a.  A  tape is created with JCL specifying UNIT=2400-3
         and no density is given.

     b.  The system allocates this tape to a 1600/6250
         drive.  It will be written at the drive's default
         density of 6250 bpi.

     c.  When running subsequent jobs calling for a data set
         on this tape, the allocation can again be either
         type of drive.  If allocated on the 800/1600 drive,
         the tape, being written at 6250 bpi, will not be
         read successfully.

If use of the IBM generic unit specifications (e.g. 3400-5,
2400-4) is  necessary the user must explicitly specify DEN=3 for
1600 bpi to insure that the tape will be written at a density
acceptable  on either type of tape drive .   The DEN parameter is
ignored on  input except for 7-track tapes.

NCC-IBM USER'S GUIDE                                     JAN  1988

No volume serial parameter (VOL=SER=vvvvvv) is given on the
sample DD statement above.  In a standard IBM environment this
would call for a mount of volume serial SCRTCH.   However, with
the Tape Management System, a IMS scratch volume is requested.
Deallocation messages as well as the IMS message in the job's JES
log will indicate the actual volume serial number of the tape
assigned.  A volume serial reference must be included on a job
allocating an existing tape data set that has not been
catalogued.   For example, for a tape data set on the fifth file
of a standard labelled tape 021111, recorded at 6250 bpi, the
following would be all of the information necessary to process
the data set:


DCB - There are many possible subparameters to describe the  data
control block (DCB).  Only the most common are described below.

RECFM - specifies the record format.  The first character may be
any one of the following:

     U - Undefined record format.  LRECL should not be
     coded.  BLKSIZE must be greater than or equal to the
     largest block written.  When reading from or writing
     to this data set, entire blocks are referenced.  They
     may be any size up to the BLKSIZE.  It may only be
     used in conjunction with A and M.

     V - Variable length records.  LRECL must be the length
     of the largest logical record plus four bytes for a Record
     Descriptor Word.  BLKSIZE must be at least as large
     as LRECL plus four bytes for a Block Descriptor Word.

     F - Fixed length records .   BLKSIZE must be a multiple
     of LRECL, the multiple of one for unblocked records.

     D - Variable length ASCII tape records.

The following characters may be used in any combination after the
first character:

     B - Blocked records (only used with F or V in the first
     character).  Blocking forms a physical block made up of
     a number of logical records .  An inter-record gap is
     recorded between each block written; therefore, a block is
     recognizable by the hardware device.

     S - When used with F (FBS or FS) this indicates fixed,
     standard format.  This is acceptable only for sequential
     data sets and indicates that there are not truncated

NCC-IBM USER'S GUIDE                                     JAN  1988

     blocks (blocks less than BLKSIZE) with the possible
     exception of the last.  For VS or VBS this indicates
     spanned records and implies that logical records
     can be spanned across block boundaries.  Exclusive
     to U or D.

     A - ASA carriage control characters are in the first
     byte of each record.  Used to control the printer.
     Cannot be used with M.

     M - Machine carriage control characters are in the
     first byte of each record.  Also used to control the
     printer.   Cannot be used with A. Cannot be used with
     ASCII tapes.

LRECL - This is used to specify the length of the logical record
for fixed or variable-length records .

BLKSIZE - The specification of the size of each physical block in
bytes.  The block size may range from 18 to 32,760 bytes for
tape .

DEN - This indicates the density at which the tape is to be
written.  If omitted, the density defaults to the highest density
permitted on the drive allocated, 6250 for the 1600/6250 drives
and 1600 for the 800/1600 drives.  When reading a 9-track tape,
the DEN subparameter is ignored since the drive can determine the
density.  The user should use care, however, with the UNIT
specification (see above).  The valid settings for the DCB
parameter's DEN= subparameter are listed in Figure 4-4.
    DEN=0      200     TAPE7, 2400-2
    DEN=1      556     TAPE7, 2400-2
    DEN=2      800     TAPE7, 2400-2, TAPES, 2400-4
    DEN=3     1600     TAPE, TAPE16, TAPE9, TAPE62, 2400, 3400-5
    DEN=4     6250     TAPE, TAPE16, TAPE9, TAPE62, 2400, 3400-5

                   Figure 4-4.  Values for DEN
LABEL - This parameter is only relevant for tape data sets .
options are listed below:

     LABEL=(n) where n is a number specifies the file
     location of the data set.

     LABEL=(,NL) indicates no label.  The system will check
     to be sure that no label exists on the tape.  Each
     individual file represents a separate data set.

NCC-IBM USER'S GUIDE                                     JAN 1988

     LABEL=(,SL) is the standard IBM tape label.  The label
     retains information on the DCB attributes, volume
     serial number, expiration date, and the last 17
     characters of the data set name.   Each data set file is
     preceded by a standard header and followed by a
     standard trailer so that the data set in effect takes
     three files.   In the example above the data file (the
     second standard file) is actually the fifth real file
     on the tape =

          First file - first standard header
          Second file - first data set
          Third file - first standard trailer
          Fourth file - second standard header
          Fifth file - second data set
          Sixth file - second standard trailer

     A double tape mark is a standard IBM convention
     indicating end of volume.

     LABEL=(,AL) is the American National Standard label;
     Sperry standard label tapes may be processed with this
     label type.  The DCB subparameter OPTCD=2 must also be
     specified.  An output data set going to this type of
     tape must have a block size between 18 and 2048.

     LABEL=(,BLP) indicates bypass label processing.  No
     label checking will be attempted.  Labels will be
     treated  as data.  In the example above, if BLP were
     specified, the fifth file  would have to be indicated to
     position the tape to the same data file; i.e.,
     LABEL=(5,BLP).   If a tape  is to be read with NL or BLP
     specified, all DCB attributes other than density must
     be supplied since that information cannot be obtained
     from the label.

     LABEL=(,,,OUT) shows use of the fourth positional
     parameter which can be either of the keywords IN or
     OUT.  Coding IN insures that the file can only be
     opened for input; OUT indicates output only.  These
     subparameters are useful primarily for programs written

NCC-IBM USER'S GUIDE                                     JAN  1988

     in FORTRAN which opens data sets for INOUT or OUTIN,
     depending on whether a READ or WRITE is first issued on
     the data set.  If the subparameter IN is not specified,
     the system will always insist that a write-enable ring
     be in a tape, requiring operator intervention to insert
     the ring.  If applicable, RACF authority of UPDATE or
     the WRITE password, would be required even though the
     program only needs to use the data set as input.

     LABEL=(EXPDT = 84340 ) provides the expiration date which
     prevents the tape from being returned to the IMS
     scratch pool before the time specified.  TMS has a
     number of expiration dates with special meanings which
     are explained in Section  RETPD=nnn can be
     used for the same purpose where nnn is the number of
     days to retain the tape from the current date.  EXPDT
     and RETPD are keyword parameters so LABEL=EXPDT=84340
     is valid and need not be coded LABEL=(,,,,EXPDT=84340).    Read or Input Only Files

Some programs, often FORTRAN programs, open a data set for
input/output even though only read operations are to be executed
against the data set.  If the input data _et is Resource Access
Control Facility (RACF) protected with a universal access code of
READ (UACC=READ), the application will ABEND with a RACF-
generated code (S913) due to the output or write capability.
This characteristic can be overridden in either TSO or batch

In TSO, use the INPUT option on the allocate to indicate that the
data set is to be read only.  An example follows:

     FREE F(FT06F001 FT07F001)
     ALLOC F(FT06F001) DA(*)
     ALLOC FCFT07F001) DA(faaaa.filename') INPUT
     CALL 'iiiaaaa.loadlib(membername)'
     FREE F(FT06F001 FT07F001)

NCC-IBM USER'S GUIDE                                     JAN 1988

In batch, use the LABEL subparameter IN.  An example follows:

     //iii   JOB (aaaauuuuu,Bxxx),READONLY,PRTY=4,TIME=(,3),
     //      NOTIFY=iii,MSGCLASS=P
     //*     CNTL(READONLY)
     /*JOBPARM      FORMS=8381
     //STEP1    EXEC PGM=GKM01
     //STEPLIB  DD DISP=SHR,DSN=filename.loadlib
     //FT06F001 DD SYSOUT=*
     //FT07F001 DD DISP=SHR,DSN=aaaa.filename,LABEL=(,,,IN)
4.1.4    OUTPUT JCL Statement

The  OUTPUT JCL statement is used to specify processing options
for  a system output (SYSOUT) data set.  These processing options
are  used only when the OUTPUT JCL statement is explicitly or
implicitly referenced by a SYSOUT DD statement.  JES combines the
options for this OUTPUT JCL statement with the options from the
referencing DD statement.

When the OUTPUT JCL is placed before the first EXEC JCL
statement, it is considered to be a job level statement which may
apply to all of a job's DD statements.  OUTPUT JCL statements
specified after an EXEC JCL statement are considered to be step
level statements.  Step level statements replace the job level
statements and apply to that step's DD statements, unless
specifically from a DD statement.  There *s no limit to the
number of step or job level OUTPUT JCL statements, and they can
be  intermixed in the same  jobstream.  If any OUTPUT and/or DD
keywords are not specified, JES2 establishes the defaults.

Since OUTPUT JCL is a functional replacement for the /^OUTPUT JCL
statement, on a data set basis, when an OUTPUT JCL statement is
used, the /^OUTPUT JES statement is not applied.

OUTPUT JCL statements are  useful in processing the output of one
SYSOUT data set in several ways.  For example, a SYSOUT data set
can  be sent to a distant site for printing, as shown in statement
OUT1, while it is also printed locally, as shown in statement
OUT2 :

NCC-IBM USER'S GUIDE                                     JAN  1988

    //DS   DD     SYSOUT=C,OUTPUT=(*. OUT1 ,*. OUT2 )

The OUTPUT JCL statement consists of the characters // in Columns
1  and 2 and 4 fields:  name, operation (OUTPUT), parameter, and

The rules  for coding the OUTPUT JCL statement are as follows:

    a.  Code a name in the name field of every OUTPUT JCL statement
        with these restrictions:

        o   Each OUTPUT JCL name must be unique within a job.
        o   The name must begin in Column 3.
        o   The name is 1-8 alphanumeric or national characters.
        o   The first character must be alphabetic or national.
        o   The name must be follow by at least one blank.

    b.  The operation field consists of the characters OUTPUT
        and must be preceded and followed by at least one blank.
        It can begin in any column.

    c.  The OUTPUT JCL statement contains only keyword parameters.
        All parameters are optional; however, the parameter field
        cannot be blank.   Valid keyword parameters, coded in any
        order in the parameter field, are described below.    Parameters

BURST=Y (or N)

Y indicates that the printed output from a 3800 printer is to be
burst into separate sheets.  N indicates that the printed output
is to be in continuous, fanfold mode.  N is the default.


The name of a character arrangement table for a 3800 printer.
Each name  is 1-4 alphanumeric or national characters; from 1-4
names can  be coded.  To specify more than one name, use


A value that specifies the maximum number of lines or cards
contained  in a logical page.  The default is the value specified
in the JES2 initialization parameter for the device.  For
printers,  it can be 0-32767; for punches it can be 1-32767.

NCC-IBM USER'S GUIDE                                     JAN  1988


A value that specifies the number of logical pages to be printed
before the next checkpoint is taken.  The default is the value
specified in the JES2 initialization parameter for the device.
Can be 1-32767.


Identifies the output class for the data set.  The class is one
character: A-Z or 1-9.  See Section 4.3.1 of this guide for valid
classes at NCC.


PROGRAM indicates that each logical record in the data set begins
with a carriage control character.

SINGLE indicates single spacing.

DOUBLE indicates double spacing.

TRIPLE indicates triple spacing.


COPIES specifies how many copies of the SYSOUT data set are to be
printed.   The printed output is in page sequence for each copy.
nnn is 1-3 decimal number from  1-255.


COPIES=(,(group-value[ ,group-value ]...))

Specifies how many copies of each page are to be printed before
the next page is printed.  Each group-value is 1-3 decimal
numbers from 1-255 in a JES2 system.

A maximum of 8 group-values can be coded.  Their sum must not
exceed 255.   The total copies of each page equals the sum of the
group values.  Note that this subparameter is valid only for 3800
output, and that it overrides an nnn subparameter, if coded.

NCC-IBJ1 USER'S GUIDE                                     JAN  1988


Allows the user to implicitly reference SYSOUT DD statements
within the job.  DEFAULT=YES means that the OUTPUT verb's
keywords are to be used as the defaults for all the eligible
SYSOUT DD statements within the OUTPUT verb's scope.  This scope
is either the step or the job, depending on the placement of  the
OUTPUT verb.  An eligible SYSOUT DD statement is one which does
not explicitly reference an OUTPUT verb.

DEFAULT=NO means that the OUTPUT statements apply only to the DD
statements which explicitly reference the statement.  The default


The OUTPUT JCL DEST keyword allows only one destination.
However, multiple OUTPUT JCL statements can implicitly reference
the same DD statement and the DD statement may explicitly
reference up to 128 OUTPUT JCL statements.


An alphanumeric value indicating the data set forms control or
carriage specifications (1-M characters).


Identifies the forms overlay to be used in printing the SYSOUT
data set on a 3800 printer, and, optionally, to specify the
number of copies on which the forms overlay is to be printed.



Identifies the forms overlay that the operator is to insert into
the printer before printing begins.  The name is 1-4 alphanumeric
or national characters.



Specifies the number (0-255) of copies that JES is to flash with
the overlay, beginning with first copy printed.  A code of zero
will flash no copies.


Identifies the forms on which the SYSOUT data set is to be

NCC-IBM USER'S GUIDE                                     JAN 1988

printed or punched.  form-name is 1-8 alphanumeric or national



Indicates a standard form.


Specifies that the SYSOUT data set belongs to an output group.
The data sets in an output group are processed together in the
same location and time.  Data sets to be grouped should have
similar characteristics: the same output class, destination,
process mode, and external writer name.  output-group is 1-8
alphanumeric characters and is selected by the programmer to
define an output group for the job.   The name is not


Sets the left margin for output on a 3211 printer with the
indexing feature.  The width of the print line is reduced by the
INDEX parameter value.  nn specifies how many print positions the
left margin is to be indented (1-31).  n= 1 indicates flush-left;
n=2 through n=31 indent the print line by n-1 positions.


Processes the all of the jobs system-managed data sets.


Processes all job control statements in the input stream, that
is, all JCL statements and JES2 or JESS statements, plus all
procedure statements from any in-stream or cataloged procedures a
job step calls, plus all messages about job control statements.



Processes the job's hard copy log, which contains the JES and
operator messages about the job's processing:  allocation of
devices and volumes, execution and termination of job steps and
the job, and disposition of data sets.



NCC-IBM USER'S GUIDE                                     JAN 1988

Processes any system messages for this job.


Sets the right margin for output on a 3211 printer with the
indexing feature.  JES2 ignores the LINDEX parameter if the
printer is not a 3211 with the indexing feature.


Specifies the maximum number of lines JES2 is to print on each
output page.  nnn is 0-255.  LINECT=0 keeps JES2 from starting a
new page when the number of lines exceeds the JES2 initialization


Specifies a copy-modification module that tells JES how to print
the SYSOUT data set on a 3800 printer.  The module can specify
legends, column headings, and where and on which copies the data
is to be printed.  module-name is 1-4 alphanumeric or national
characters and is in SYS 1.IMAGELIB.


Identifies the universal character set.  The code is 1-4
alphanumeric or national characters.   Refer to the OS/MVS
Extended Architecture JCI Reference Manual (GC28-1352) for more


Names an external writer to process the SYSOUT data set rather
than JES.  An external writer is an IBM- or installation-written
program (1-8 alphanumeric characters).  INTRDR tells JES that the
SYSOUT data set is to be processed by the internal reader.

4.1. 4. 2    Implicit Mode

OUTPUT JCL can be referenced by the DD SYSOUT statement in
implicit mode when the OUTPUT keyword DEFAULT is used.  The
DEFAULT=YES keyword means that the OUTPUT keywords are to be used
as the defaults for all the eligible SYSOUT DD statements within
the step or the job, depending on the placement of OUTPUT JCL.
An example follows =

    //EXAMPLE 1 JOB(aaaauuuux,bin),name,MSGCLASS = A
    //R1       OUTPUT DEFAULT=Y,DEST=R1,FORMS=6371
    //R2       OUTPUT DEFAULT=Y,DEST=R2,FORMS=6151
    //STEP1    EXEC     PGM=WRITE

NCC-IBM USER'S GUIDE                                     JAN  1988

    //R3       OUTPUT   DEFAULT=Y,DEST=R3,FORMS=61 1 1
    //DD1      DD       SYSOUT=A
    //STEP2    EXEC     PGM=SCAN
    //DD2      DD       SYSOUT=(A, ,8381 )

STEP1 has an OUTPUT JCL statement with DEFAULT=YES, and
therefore, the R1 and R2 job level default OUTPUT  statements  do
not apply to the DD1 statement.  DD1 will obtain  default
information from the R3 OUTPUT statement and will  have the
following characteristics =


STEP2 does not have OUTPUT JCL.  Thus the job level OUTPUT JCL
with DEFAULT = Y will provide all missing keywords .   Two copies of
DD2 SYSOUT will be created with the following characteristics:

     COPY1=   CLASS=A,FORHS=8381,DEST=R1
     COPY2:   CLASS=A,FORHS=8381,DEST=R2    Explicit Mode

OUTPUT JCL is explicitly referenced when the OUTPUT keyword is
coded on the SYSOUT DD statement.  The format of this new keyword
is as follows:

    //DDname DD SYSOUT = n,OUTPUT=(*.name, . .  . )
                                 (*.step.name, . .  . )
                                 (*.proc.step.name, . . . )

The OUTPUT keyword of the DD statement specifies 1-128 OUTPUT
verb names which supply the output characteristics for the DD
statement.  The referenced OUTPUT statements must  occur in the
jobstream before the DD statement.

Often the same set of job control statements is used repeatedly
with little or no change (for example, to specify compilation,
link-editing, and execution of programs).  To save programming
time and to reduce the possibility of error, standard job step
definitions can be prepared and placed (or catalogued) in a
partitioned data set known as a procedure library.  A set of  job
control statements that has been assigned a name and stored in a
procedure library is called a catalogued procedure, or PROC.
Catalogued procedures are placed as members in one of four
partitioned data sets =

NCC-IBM USER'S GUIDE                                     JAN  1988

     a.  PROCLIB+ - private procedure libraries (see Section

     b.  SYS1.PROCLIB - IBM system software only.

     c.  SYS1.PRODPROC - language and vendor applications;
         maintained by Systems.

     d.  SYS1.USERPROC and SYS2.USERPROC - user procedures
         (see Section 4.2.7).

An alternative type of procedure is an in-stream procedure.   An
in-stream procedure is a set of job control statements placed in
the jobstream immediately before the first EXEC statement.  This
can be used any number of times during a job by naming the
procedure on an EXEC statement.   In-stream procedures may be  used
to test JCI before saving it as a catalogued procedure.

Catalogued and in-stream procedures may have symbolic parameters
coded in the JCL statements.   Symbolic parameters provide a means
of generalizing JCL and may be assigned default values within the
procedure.  These values may be assigned or overridden in the
EXEC statement invoking the procedure.  The EXEC statement can be
followed with DD statements to override or supplement the ones
already contained in the procedure.

A procedure may consist of one or more steps.  Each procedure
step consists of an EXEC statement and DD statements.  The EXEC
statement identifies the program to be executed; the DD
statements define the data sets to be used by the program.
Procedures can contain all but the following JCL statements:

     a.  EXEC statements invoking procedures.  (A procedure
         cannot invoke another procedure.)

     b.  JOB, delimiter (/*), or null (//) statements.

     c.  JOBLIB DD statements.

     d.  DD * or DATA statements.

     e.  Any JES2 control statements.

Users may find more detailed information on procedures in the IBM
OS/VS2 MVS JCL Reference Manual (Order Number GC28-0692) and  in
the book by Gary DeWard Brown entitled SYSTEM/370 JOB CONTROL
LANGUAGE published by Wiley-Interscience,  John Wiley £ Sons,  605
Third Avenue, New York, NY  10016.

NCC-IBM USER'S GUIDE                                     JAN  1988

4.2.1    Catalogued Procedures

Catalogued procedures are kept in one of the procedure libraries
listed above.  The libraries are concatenated in the order given
and searched in that order.

A catalogued procedure begins with a PROC statement and is
followed by JCL statements.   The PROC statement is optional if
symbolic parameters are not assigned default values.   The PROC
statement is as follows:

    //procedure-name PROC symbolic-parameters comments

The procedure name must be unique within the procedure library.
The procedure-name is 1 to 8 alphanumeric characters (A to Z,
0-9) or national (<3 * #) characters.  The first character must
not be numeric.

Section 4.2.6 contains a sample procedure.

4.2.2    In-Stream Procedures

The first statement of the in-stream procedure must be a PROC
statement with the procedure name in the label position (starting
in column three).  The last statement must be a PEND statement.
The procedure itself must appear in the jobstream before the
first EXEC statement calling for that procedure.  Up to 15 in-
stream procedures  can be included in a single job.  Each in-
stream procedure may be invoked several times within the job.

4.2.3    Coding Symbolic Parameters

A symbolic parameter stands  as a symbol for a character string on
an EXEC or DD statement in a procedure.  Symbolic parameters
allow the user to  code variable information in the operand field
of a procedure EXEC statement or DD statement.  The value to  be
assumed by a symbolic parameter may be coded on the EXEC
statement that calls the procedure and is in effect only during
the current execution of the procedure.

In the body of a procedure,  a reference to a symbolic parameter
is made by coding  an ampersand (£) followed by the parameter
name.  When the procedure is executed, the value assigned to  the
symbolic parameter will be substituted for such a reference
before analysis of the JCL statement.  Any symbolic parameter
referenced in the  body of a procedure must be assigned a value
before the procedure is executed.

There are two ways that a symbolic parameter can be assigned  a


NCC-IBM USER'S GUIDE                                     JAN  1988


     a.  A value can be assigned to the symbolic parameter
         on the EXEC statement that calls the procedure.

     b.  The PROG statement, which may appear as the first
         statement in a catalogued procedure and must appear
         as the first statement in an in-stream procedure,
         may be used to assign a default value to the
         parameter.  Any default assigned to a symbolic
         parameter on the PROC statement is overridden when
         a value is assigned to the same symbolic parameter
         on the EXEC statement that calls the procedure.

The ampersand is only prefixed to a symbolic parameter in the
body of a procedure; it is not used in the PROC or EXEC

Coding the following on the EXEC or PROC statement (see b. above)
assigns a value to a symbolic parameter:


For example, if the reference SNUMBER (calling for the value  of
the symbolic parameter NUMBER) appears on a DD statement in the
procedure, the user assigns a default value to the parameter  by
coding NUMBER=value on the PROC statement.  If the parameter  is
to be in effect only during the current execution, the user
should code NUMBER=value on the EXEC statement calling the
procedure.  The user must not assign a value to a parameter that
is not referenced within the body of a procedure; a JCL error
results.  If the value contains special characters, it must be
enclosed in apostrophes.  The enclosing apostrophes are not
considered part of the value.
M . 2 . M    Nullifying Symbolic Parameters

To nullify a symbolic parameter is to assign it a value
equivalent to a character string of zero length (a null string).
For EXEC and DD statements operands of the form "keyword=value",
a value of a null string causes the Keyword to be ignored.

Coding the following on the EXEC or PROC statement nullifies a
symbolic parameter:


No value follows the equal sign.

The following sample illustrates an in-stream procedure set up  by

NCC-IBM USER'S GUIDE                                     JAN  1988

user UK (account AAAA) to print specified data sets with prefix
IJKAAAA. In this example member MEMB from data set IJKAAAA.LIB
will be printed to class A.

     // . . .  JOB ...
     //       PEND
M . 2 . 5    Overriding Catalogued Procedures    Overriding EXEC Statement Parameters

When executing a procedure, either catalogued or in-stream, users
have the option of overriding, adding, or nullifying parameters
coded on EXEC statements contained within the procedure.  Any
desired changes can be coded on the EXEC statement which calls
the procedure.  Overriding the PGM parameter is not permitted.

The following format is used to change or add an EXEC statement
parameter =


For example, if the EXEC statement of the first step, STEP1, on a
catalogued procedure specifies PARM=ABC, the user may change the
FARM value to CBA by specifying the following when calling the
procedure =


To  nullify this same FARM, he may code the following:


All parameters to be overridden, added, or nullified for one
procedure step must be coded before those of the next step.

For example, given a procedure PROCA with three steps, STEP1,
STEP2, and STEPS, the user may wish to override the region size
in  the first, the condition code in the second, and the parameter
being passed in the third; the EXEC card invoking the procedure
must be as follows:

NCC-IBM USER'S GUIDE                                     JAN 1988

     //  COND..STEP2=' (4,GT,STEP1 ) ' ,
     //  PARM.STEP3='1,2,3*

The procstepname (e.g.,  STEP1, STEP2, STEPS above) may be omitted
when coding an EXEC statement parameter.  In that case, the
procedure is modified as follows =

     a.  If the PARM parameter is coded, it applies only to
         the first procedure step.   If a PARM parameter
         appears in a later EXEC statement, it is nullified.

     b.  If the TIME parameter is coded, it applies to the
         total procedure.  If the TIME parameter appears on
         any of the EXEC statements in the procedure, it is

     c.  If any other parameter is  coded, it applies to
         every step in the procedure.  If the parameter
         appears on an EXEC statement, it is overridden. If
         the parameter does not appear on an EXEC statement,
         it is added.    Overriding DD Statement Parameters

Parameters coded on DD statements in a catalogued procedure can
be overridden, added, or nullified at execution time.  To
override a parameter on a DD statement within a procedure, it is
necessary to include a DD statement following the EXEC statement
that calls the procedure.  The DDNAME of this DD statement must
identify the DD statement that contains the parameter to be
overridden and the procedure step in which this DD statement
appears.  One DD statement is used to override, add, and nullify
any parameters contained on the same DD statement in the
procedure.  The format required for a DD statement following the
EXEC statement is as follows;

     //procstepname.ddname DD parameter=value

In the operand field of this DD statement, the user can code a
parameter to be overridden and its new value, a mutually
exclusive parameter to replace an existing parameter, or an
additional parameter.  Furthermore, a DD statement parameter can
be nullified by omitting the value following the equal sign.

When overriding parameters on more than one DD statement within a
procedure, the overriding DD statements must be in the same order
in the input stream as the corresponding DD statements within the
procedure.  (See the example in Section

NCC-IBM USER'S GUIDE                                     JAN  1988    Adding DD Statements to a Procedure

Additional DD statements can be added to a procedure at the time
that the procedure is executed, for the duration of that
execution.  The additional DD statements must follow the EXEC
statement that calls the procedure and any overriding DD
statements for that procedure step.

The format of a DD statement to be added to a procedure step is
as follows:

     //procstepname.ddname DD parameters

As soon as the first DD name is encountered in the jobstream for
which there  is no matching DD name in the procedure, the system
assumes that all remaining DD cards for the step are additions,
not overrides.  The effect will be that such misplaced overrides
will be ignored.

M . 2 . 6    Sample Procedure

The example  below shows procedure COBUCL available at NCC.   The
first step compiles COBOL source programs and the second step
links the compiled program into a load module.

NCC-IBM USER'S GUIDE                                     JAN  1988

COBUCL:  Procedure to compile and link a COBOL source

     //*                               LINKAGE DEFAULTS
     // SPACE=(3120,(100,10))
     //SYSUT1   DD UNIT = SYSDA,SPACE=(3120, (200,200) )
     //SYSUT2   DD UNIT = SYSDA,SPACE=(3120, (200,200) )
     //SYSUT3   DD UNIT = SYSDA,SPACE=(3120, (200,200) )
     //SYSUT4   DD UNIT = SYSDA,SPACE=(3120, (200, 200) )
     //         DD DDNAME=SYSIN
     // SPACE=(CYL,(2,2,1))
     // SPACE=(3120,(200,200))

Notes:  (see numbers in parentheses)

(1)  The symbolic CPARM is shown defaulting to a null string.

(2)  DDNAME SYSLIN is conventionally used  as  output of the compiler
     step and input to the Linkage Editor  step.

(3)  See Section 5.3.9 on the Linkage Editor  for a brief
     explanation of the FARM field or refer to the IBM
     manual MVS Extended Architecture Linkage Editor and
     Loader (GC26-4011).

(4)  Condition code tests are often difficult to interpret.
     The meaning of the test, COND=(4,LT,COB) should be read
     as follows =

     "If 4 is less than the condition code in step COB, then
      bypass this step."

     Consequently, any condition code greater than 4 in step
     COB will result in not executing step LKED.

NCC-IBM USER'S GUIDE                                     JAN 1988

(5)  DDNAME SYSLMOD is always used for the output of the
     Linkage Editor; i.e., a load module.  This procedure
     sets up a temporary data set to be passed to a
     subsequent step.

In order to demonstrate the use of this procedure, consider the
following situation;

     a.  The user has  a COBOL source program called COBPROG
         in library IIIAAAA.ASM.

     b.  A load module called PROGRAM with an alias of
         COBPROG is to be stored in IIIAAAA.LOAD.

     c.  The Linkage Editor is to reject a compilation with
         a condition code of 4 or more.

The necessary JCL to invoke this procedure is as follows:

     //III   JOB...
(4)  //LKED.SYSIN   DD *

Notes:   (see numbers in parentheses)

(1)  The user should remember that when overriding keywords used
     on an EXEC statement, the keyword precedes the step name.
     In cases where a  DDNAME is overridden or added to a jobstream,
     the step name comes first.

(2)  Most of the compiler procedures do not indicate SYSIN, but
     it must be present for the definition of the source module.

(3)  If overriding a DD statement allocating a new data set, the
     user should be sure to nullify any parameters necessary.
     UNIT and SPACE should not be indicated for an existing
     catalogued data set.

(4)  Because of the deferred DD definition under DDNAME SYSLIN
     in the procedure, this has the effect of concatenating the
     control cards which follow with data set fiOBJMOD.

(5)  The control cards establish the alias and load module
     name.  If the alias had not been necessary, LKED.SYSIN DD *
     and the control cards could be deleted and the member name
     PROGRAM specified explicitly on the LKED.SYSLMOD statement.

NCC-IBM USER'S GUIDE                                     JAN  1988

4.2.7    User Procedures

In addition to the catalogued procedures available at the NCC,
users may find that they have sets of JCL statements which are
used repeatedly.  To avoid long JCL streams and lessen the
possibility of error, standard sets of JCL may be stored
(catalogued) in the procedure libraries maintained for users.
The permanent procedure libraries for EPA users are named:

     SYS1.USERPROC and

SYS1.USERPROC will be updated daily at 10=00 AM, 2=00 PM, and
5=00 PM (EST or EOT) using members from the temporary libraries.
SYS1.USERPROC is write-protected to prevent users from updating
their PROC's directly into the library.  In order to add or
replace members in this library, users must first place their JCL
routines in one of two update libraries.   The user may make
updates and additions to SYS2.USERPROC directly since there is no
protection on this procedure library.  Requests for special
updates should be directed to User Support.  The names of the
temporary libraries in which users place their 80-character JCL
procedures are:

     SYS2.PROCADD  for adding new procedures
     SYS2.PROCRPL  for replacing existing procedures.

All JCL procedures at NCC must conform to certain conventions.
The first JCL statement must be a comment card formatted as

         Column         Contents

          1-3            //* (identifies JCL comments card)

           4             blank

          5-12           iii/aaaa (User-ID and account of
                         responsible person)

          13             blank

         14-71           name and phone number of responsible
                         person (free form)

         72-80           as desired

The second statement must be a PROC job control card with the
procedure name (or member name) starting in Column 3.  User
catalogued procedures not conforming to these conventions will be
periodically removed from both SYS 1 .USERPROC and SYS2.USERPROC.

NCC-IBM USER'S GUIDE                                     JAN  1988

Users can place their JCL procedures in the temporary libraries
or into SYS2.USERPROC using one of the following methods:


     EDIT procedure-name CNTL NEW
     *** build JCL in workspace
     SAVE 'SYS2.PROCADD(procedure-name)'


Replacing user catalogued procedures =

     a.  Copy SYS 1.USERPROC(procedure-name) to
         SYS2.PROCRPL(procedurename) using Option 3.3 or
         Option 2.

     b.  Use Option 2 to edit and save
         SYS2.PROCRPLCprocedure-name ) .

Adding user catalogued procedures:

     a.  Use Option 2 to add code to SYS2.PROCADD(new-
         procedure-name ) .

     b.  Save SYS2.PROCADD(new-procedure-name).

     Note:  NCC saves user PROC's until NCC-IBM User Support
            runs a job that transfers the new user PROC to


     //useridXX JOB ...
     ./  ADD NAME=procedure-name
     //* iii/aaaa J.SMITH (202)999-9999
     //procedure-name PROC
     Place Balance of JCL Procedure  Here  (Remove all /*
     ./  ENDUP

A word of caution:  Each new procedure must be uniquely named.

NCC-IBM USER'S GUIDE                                     JAN  1988

Semiannually, users will be mailed a list of all catalogued
procedures by user identification code.  He may respond by
indicating the procedures to be retained as instructed and return
the list to User Support.  A more desirable method of responding
would be to execute RETPROC CLIST by entering the following at
the TSO READY prompt'•


Only those procedures  requested to be retained will be in the new
SYS1.USERPROC.   Consequently, failure to respond could result in
the deletion of currently used procedures.  If this happens,
contact User Support for assistance (see Appendix A for telephone
4.2.8    Private Procedure Libraries

PROCLIB+ is an IBM facility which allows users to create and
maintain their own private procedure libraries.  A partitioned
data set is used as a private PROCLIB.  This PDS must reside on
one of the permanently mounted disk packs and must be catalogued.
All standard rules for the creation and :.. ..intenance of a PDS
apply to private user procedure libraries with two exceptions =

     a.   The blocksize must be 3120 (as for SYS 1.PRODPROC).

     b.   The logical record length must be 80.

An example of the JCL needed to allocate a user  PROCLIB is as

     //            DCB=(RECFM=FB,BLKSIZE=3120,LRECL=80),
     s/            DSNAHE=iiiaaaa.procname,
     //            SPACE=(TRK,(pri,sec,dir))

Users can call a procedure out of a private library by inserting
a PROCLIB DD statement after the job card but before the first
EXEC card.  The PROCLIB DD statement defines a private library to
be searched before the system PROCLIB's such as  SYS 1.PROCLIB,
SYS1.PRODPROC, or SYS 1 .USERPROC.   The search will be for
procedure names indicated in all EXEC cards contained in the job
stream.   This is similar to the way a JOBLIB is  searched for a
program.   The DD name "PROCLIB" must be used on  the DD statement
identifying the user PROCLIB's; all other DD names are invalid.
Users may concatenate several data sets within the PROCLIB DD
card.   These will be searched in the order in which they are

NCC-IBM USER'S GUIDE                                     JAN  1988

Following is an example of a PROCLIB DD card with two libraries
to be searched for procedures =

     //. . .JOB. . .
     //        DD DISP=SHR,DSNAME=aaaa.PROCLIB2
     //stepname EXEC procname

M . 3. 1    SYSOUT Classes

The SYSOUT parameter is used to identify the unit record device
upon which a data set is to be printed or punched.  This
parameter on the DD statement specifies that the operating system
is  to spool the data set to disk automatically, from which it
will be output to the appropriate device when resources are
available.  This is done so that the computer's printing and
punching resources may be shared among many jobs which have
already "completed execution and also so that jobs may be executed
independently of the availability of SYSOUT resources.

An example of the SYSOUT parameter follows, using the IEBPTPCH
utility to write the contents of a data set to the line printer:

     //SYSUT2   DD SYSOUT=class

NCC-IBM USER'S GUIDE                                     JAN 1988

The most commonly used classes at the NCC are =

Class           Meaning

A               Printed output on standard system printer with
                single-part, unlined 14-7/8" x 11" white paper
                printed at 6 lines/inch.

B               Punched output on standard 80-column cards.

D               Printed with upper/lower case (Washington
                Information Center N3 only).

F               Printed output on the IBM 3800 laser printer
                (local to RTF only, not currently available at
                most remote sites).

J               Printed output on special forms for individual
                data sets. Format is SYSOUT=(J,,form number).
                Defaults to upper and lower case on unlined
                14-7/8" x 11" paper at 6 lines/inch.  (Special
                form 6311).

M               Printed output greater than 50,000 lines.

Z               Dummy class.  Output will not be printed.

The JES2 system will automatically assign the DCB and other DD

4.3.2    Use of the 3800 Laser Printer

The 3800 prints up to 10,000 eleven-inch pages per hour by a
xerographic process similar to that used by the common office
copier.  It prints only on continuous, one-part forms in a
variety of sizes, and can optionally trim off the sprocket feed
holes along the edge of the forms and separate ("burst") the
forms into individual sheets after printing.  The print
characters include all upper and lower case letters, numbers, and
all special characters and punctuation marks printed by the
standard impact printers at the data center and remote terminal
sites.  IBM programming manuals for the 3800 printer describe how
to make use of additional advanced features such as creating
custom print forms, creating and using unusual character sets,
and varying the character and line spacing of printouts.  These
include Concepts of the IBM 3800 Printing Subsystem , Introducing
the IBM 3800 Printing Subsystem and Its Programming, and IBM 3800
Printing Subsystem Programmer's Guide.

The on-line usage aid data sets containing local IBM 3800 usage

NCC-IBM USER'S GUIDE                                     JAN  1988

information are JUSD.IBM3800.DATA and JUSD.CHAR3800.DOC.

The 3800 has the following limitations:

     a.  It cannot print self-stick labels, because the
         print form must bend through a sharp curve inside
         the printer which tends to separate the labels from
         their backing sheet.

     b.  It cannot print on multi-part paper, because the
         lithographic  nature of the printing process can
         transfer an image to only one paper surface at a
         time. But the user may request more than one copy.

     c.  It cannot overprint lines (usually used to simulate
         bold-face type or to underline characters), because
         the print form must move smoothly and continuously
         through the machine during printing much as in a
         newsprint press.

The 3800 does detect an attempt to overprint lines for
underlining text and automatically prints the desired
underlining, but other effects usually achieved by overprinting
must be accomplished by programming to make use of advanced
features of the 3800.

Finally, the 3800 printer  cannot print within one-half inch of
the top or bottom of each  page, because this space is reserved
for starting and stopping  the print forms whenever the paper
movement must be interrupted during printing.  Since standard
print output at NCC-IBM may contain 61 lines per page at 6 lines
per inch, lines 58 through 61  are printed on the 3800 printer at
8 lines per inch to accommodate them within the printable  10
inches of the 11-inch  standard page length (i.e. 57 lines at 6
lines/in = 9-1/2" + 4  lines at 8 lines/in. = 1x2").    Forms for the 3800

The 3800 printer can print on all of the one-part forms (except
for self-stick labels) used on the other data center printers and
documented in Section  4.3.7.  It can also print on one-part forms
in a variety of other  sizes and weights as described in the first
two IBM publications listed in Section 4.3.2.  Forms in common
use at NCC-IBM include:

     1)  6111 — the standard 14.875" by 11" form with upper
         case printing used by default for all printouts
         which do not  request a specific form.

     2)  6151 — 9.5"  by 11" form (upper case).

                                              JAN 1988
     3)  6171 — 9.5" by 11" form punched for a 3-ring
         binder (also upper case).

NCC-IBM forms can be printed in both upper and lower case by
substituting a 3 for the second digit of the form number.  For
example, form 6371 is used to print both upper and lower case on
letter-sized, 3-hole punched paper.

Because the 3800 printer can print clearly using smaller, more
closely-spaced characters than standard computer printers, it is
possible to compress the size of a standard  14.875" by  11" print
page to fit on a 12" by 8.5" form which can  be punched  and bound
in an ordinary ring binder.  To take advantage of this  feature,
new form numbers are available.  These forms, described below,
may be used only with the 3800 printer; they are not available on
the standard NCC printers.




NCC-IBM USER'S GUIDE                                     JAN  1988

printouts are normally produced at a site other than the data
center at RTF, you should explicitly specify DEST=RMTO on all RUN
or LIST OFFLINE commands which will produce 3800 printer output.    Job Control Language

To print all output from a job on the IBM 3800 printer, specify
in the JOB card the parameter MSGCLASS=F, and specify SYSOUT=F on
every DD statement which describes a print output file.  For

  //iii JOB (acctuid,room),'prog name',MSGCLASS=F
  //stepname EXEC . . .

To print all output from a job on any non-standard form, include
in the job a JOBPARM card which specifies the form number, for
example =


To print any individual output file on the 3800 printer, it is
not necessary to change the normal JOB card or to use a JOBPARM
card.  Simply specify SYSOUT=F on the DD statement which
describes the output file.  To print the output on any non-
standard form, add the form number to the DD statement, for
example =

  //iii        JOB (acctuid,room),'prog name1
  //stepname   EXEC
  //PRINT1     DD SYSOUT=(F,,8381)
4.3.3    Use of the Laser Printers at WIG

High-quality, channel-connect printing from the IBM 3090 is
available at WIC through use of two Datagraphix 9835 laser

Form 8381  (11" x 8-1/2" with no holes) is the default output form
used on the WIC laser printer.  This form is used to print ALL
jobs with the following characteristics:

     a.  Executed on the WIC 4381 or on the IBM 3090 at NCC
         and routed to the WIC (/*ROUTE PRINT N3).

     b.  No form explicitly requested.  (i.e., no FORMS= in
         the JCL).

     c.  Explicitly request Form 8381.

NCC-IBM USER'S GUIDE                                     JAN  1988

Processing for 8381 forms will begin at 0700 hours and run until
1800 hours, Monday through Friday except during the three  1-hour
time blocks mentioned below.  Delivery of 8381 forms output to
the NIC user bins should be within one hour of the time the job
finishes executing on the IBM 3090.

Output for jobs using Form 6311 (14" x 11") will be printed on
the WIG's 3203-5 impact printer.  Because of its slower printer
speed (approximately 1,000 1pm), there is no guaranteed delivery
time .

If Form 6311 is required to be printed on the laser printer,
SYSOUT=F must be used in the JCL instead of SYSOUT=A.  This
request will then be treated as a special form and printed after
1800 hours.  Note that no new form is proposed for 14" x 11"
paper on the laser printer.  Any JCL with FORMS=6311 and SYSOUT=F
will go to the laser printer as a special form and will be
printed after 1800 hours.  Jobs only go to the impact printer
with FORI1S = 6311 and SYSOUT = A.            .

Exceptions to the WIC printing policy will be handled on a case-
by-case basis.  Users should request an exception from the WIC
Operations Supervisor at the telephone number listed in Appendix
A.  He will grant a request if the printers are not backlogged at
the time of the request or if there is no impact on the general
user community.  Users who are not satisfied with his response
may contact the Chief, Information Centers Branch at 488-5968.

All remaining forms (listed below) will be printed on the laser
printer between 0830 and 0930, again between 1130 hours and 1230
hours, again between 1530 hours and 1630 hours, and again after
1800 hours each business day.

     o  6351

     o  6371

     o  C151

     o  C171

     o  8391

     o  C181

In an effort to provide faster job output for the greatest number
of users, a limit of MO,000 lines per job is imposed during prime
shift (0700 hours - 1800 hours).  Regardless of form type, any
job that exceeds 40,000 lines will be held until 1800 hours for
processing and will be available the following business day.

                                                  JAN  1988
    Job Control Language
Three types of JCL statements are used with requests for laser
printer output.  JCL for special form requests, special print
density, number of copies, etc. should be coded on a /^OUTPUT
statement.  /*JOBPARM statements control all output for a  job.
/^OUTPUT and SYSOUT DD statements control output for particular
print files in a job.  The table below indicates the keywords
valid for the statement type.





                     SYSOUT DD

     DCB=OPTCD=J can only be specified on a SYSOUT DD statement.

     LINECT=nnn can only be specified on a /*JOBPARM statement.
     LINECT=0 allows the printing of more than 61 lines per page

     COPIES on a /*JOBPARM statement prints one complete job,
     then starts the next copy.  COPIES on a /^OUTPUT or SYSOUT
     DD statement prints all copies for that SYSOUT file, then
     continues printing the rest of the job.
    Forms for the WIC Laser Printers
The forms and print densities available at WIC are as follows:
631 1
14" x 11"
8.5" x 11"
8.5" x 11"
8.5" x 11"
8.5" x 11"
11" x 8.5"
11" x 8.5"
11" x 8.5"






The Forms Control Buffer is explained below;
                                                 JAN  1988







Max Lines/pg

for 57 lines
for 4 more
for 58 lines
for 3 more

  1 1"
  1 1"
                   1 1"
Valid Forms

6311, 6351,
6371 , C151 ,
631 1
C17 1

631 1
637 1
C17 1
6351 ,
C151 ,
6351 ,
C151 ,
6311, 6351,
6371 , C1S1,

8381, 8391,

8381, 8391,
Note that there are three forms with 3-hole punch: 6371, C171,
and 8391.
U.3.3. 3
A complete set of examples of all fonts is available at the WIG.
Some fonts are also available in different pitches  (10, 12, or  15
characters per inch).
 IBM 3820 Page Printer
Note:  The IBM 3820 hardware is being used by a segment of the
user community who have purchased the equipment.  The software is
available on the IBM 3090 for any other user groups who also
purchase the equipment.

The IBM 3820 Page Printer is an advanced function printer which
combines laser printing technology and software functions in one

High-quality output is produced on a variety of sizes and weights
of cut-sheet paper or forms appropriate to the immediate needs.

NCC-IBM USER'S GUIDE                                     JAN  1988

Duplex printing (on both sides of a sheet of paper) is a standard
capability for output on letter size paper.   Printing is produced
at burst output speeds up to 20 pages per minute.

The all-points-addressable printing capabilities together with
IBM's Advanced Function Printing software allow merging of text,
charts, graphs, drawings, logos, signatures, and electronic forms
onto any print page.

The 3820 accepts and prints pages composed of both text and
illustrations.   The Page Printer is programmed intelligently to
rearrange and print separate data elements for one page.

The ability to modify and change fonts along with the ability to
specify a variety of line spacings can contribute to the overall
readability and appearance of documents.   The amount of paper
used and the amount of paper distributed and stored is reduced.

The special graphics and images features allow a wider choice of
logos, images,  and business engineering drawings.  Time and
expense associated with manual methods are saved.

The capability to generate electronic forms  eliminates the
physical storage and costs associated with preprinted forms.
Forms with logos and other graphic images can be updated almost

The Advanced Function Printing software support for the 3820
includes the Print Services Facility, a program which manages the
communications, device control, and print resources needed to
attach the 3820.  This program interfaces directly with JES2 to
coordinate the printing of jobs on the 3820.

IBM reference manuals are listed below;

IBM 3820 Page Printer

     Introduction to the IBM 3820 Page Printer and Advanced
     Function Printing Software (G544-3080-0)

     IBM 3820 Page Printer Glossary (G544-3082-0 )

     Planning a Data Network for the IBM 3820 Page Printer

     Planning for the Location of the IBM 3820 Page Printer

     Planning for Advanced Function Printing Software for MVS

NCC-IBM USER'S GUIDE                                     JAN  1988

     Ordering Supplies for the IBM 3820 Page Printer

     Preparing an Implementation Plan for the IBM 3820 Page
     Printer (G544-3093-0)

     IBM 3820 Page Printer Operator's Guide  (S544-3080-0)

     IBM 3820 Page Printer Suick Reference (S544-3081-0)

     Advanced Function Printing Diagnosis Guide for MVS =
     IBM 3820 Page Printer (S544-3094-0)

Print Service Facility

     Print Services Facility User's Programming Guide for MVS

     Print Services Facility Messages for MVS and VSE
     (SH3S-0060-1 )

     Print Sevices Facility Data Stream Reference (SH35-0073-2)

     Print Services Facility Diagnosis  Guide and Reference for
     MVS (SY35-0063-1 )
Remote Routiner
All data sets to be printed, punched, or routed to on-line
terminals should use the SYSOUT parameter on their DD statements.
Jobs submitted via an RJE terminal will have all of their SYSOUT
data sets routed to that same RJE station by default.  Output
from jobs submitted through MYLBUR or ISO will be printed at
local NCC printers.   The user changes these defaults by the
/*ROUTE JES2 control statement or the appropriate interactive  job
submission operand (see Appendix C for a description of the
"/*ROUTE" command).   The /*ROUTE control statement, placed in  the
JCL stream between the JOB statement and first EXEC statement,
allows either PUNCH or PRINT SYSOUT data to be routed to any
valid remote destination.  Remote destinations may be specified
as RMTn where n is a one-, two-, or three-digit remote
identification.  WYLBUR users should use the DEST operand of the
RUN and LISTOFF commands.  ISO users should specify the /*ROUTE
parameter in their JCL.
Output Distribution an<
loom  Parameters
Output routed to the central site (NCC) will be processed by
Operations Support and distributed by Data Processing Support

NCC-IBM USER'S GUIDE                                     JAN  1988

Services.   Output routed to N3 will be processed and distributed
by the WIC Production Services.  At both centers, output will be
checked for proper separation and prepared for distribution.
Output will be held at either center for user retrieval when
specified  through the program job statement.  (See Appendix I.)

Washington, B.C. area users should include a ROUTE card
immediately following the JOB card directing output to N3 to
prevent default to the central site at the NCC.   The format for
the /*ROUTE card (starting in Column 1) is:

     (1)       (10)    (16)
      /*ROUTE   PRINT   N3

The NCC and the MIC support "closed shop" input/output areas  for
security purposes.  As a result, users are required to obtain a
bin ID for local output retrieval (NCC or WIC) or to utilize  a
routing code in the ROOM parameter of the program job statement.
The ROOM parameter signifies the means and the location for
output distribution.  The assigned bin ID or ROOM parameter
appears on the splash page (first page) of the output listing.

U.3.6.1    Obtaining a Bin ID

Users expecting to retrieve computer output from DPSS or WIC  I/O
Control should request a bin ID from their ADP Coordinator,
Project Officer, or Group Leader.  The responsible person should
complete a Bin Access Authoriation form (available from DPSS  or
WIC I/O Control) and return it as applicable (see Appendix A  for
addresses  and telephone numbers).

Upon receiving the form, DPSS/WIC I/O Control will assign a bin
ID and inform the ADP Coordinator.  If the bin is needed
immediately, the ADP Coordinator may obtain a bin ID from
DPSS/WIC I/O Control and inform them of the names of the persons
who will be using the bin.  The DPSS/WIC I/O Control technician
will allow those specified to retrieve output for five days or
until the  Bin Access Authoriation form arrives,  whichever is
M . 3 . 6 . 2    Room Parameter

The four-position ROOM field in the JOB card provides two types
of information.  The first position contains the route code for
disposition, and the next three positions contain the user's
assigned bin ID or the User-ID initials of the person to receive
the listings.

Valid route codes are contained in Appendix I.

NCC-IBM USER'S GUIDE                                     JAN  1988

When the route codes M or F are used, the DPSS and WIG I/O
Control will rely on the initials file maintained by the TSSMS
office for the correct name, address, and room number.  If this
information is incorrect, contact the TSSNS Office (see Appendix
A for telephone number).

Card decks produced at the NCC/WIC or at an RJE station are
identified by User-ID (from the job name) and the JES2 job
number.  Distribution of this output is determined by the EPA
registered name and the address associated with the User-ID.

Users desiring distribution to an address other than that
associated with the User-ID should contact DPSS, MIC I/O Control,
or the appropriate RJE operator to arrange distribution of their

In order to retrieve output from a specific bin, the user's name
must be included on the list of persons authorized to receive
output from that bin (see Section  In addition, each
authorized user retrieving output from a bin must pick up the
entire bin's contents, and subsequently b^ responsible for the
distribution of that output to other users assigned to that bin.

Output from bins designated for courier delivery will be routed
at least twice daily to the specified destination.  All requests
for service should be coordinated through DPSS or WIC Production
Services.   (See Appendix A for telephone numbers.)

Users who do not use the room parameter as specified above might
not receive their output, or might receive it after great delay.
For example, if user DDD uses DAVE as a room parameter the output
will be routed to WIC and mailed to AVE, if that ID exists.  If
AVE does not exist, DDD will get the output after a significant
M . 3 . 7    Retention of Output in Print Rueue

Operations executes a program nightly that examines all jobs in
the print queue.  Each data set which has been in the queue for
some  part of three working days is purged.   A working day is
defined as a day during which NCC regularly processes user jobs
(i.e., days other than Sundays or holidays).  This purging will
be  done whether or not the output has been returned.

In  general, this applies to held output, but it can affect output
routed to a remote job entry station that is down.  If this
occurs, and output may be affected, the user should contact IBM
User  Support and advise them of the situation.  They can have the
output rerouted to the central site and mailed to the appropriate
users .

NCC-IBM USER'S GUIDE                                     JAN  1988

To prevent output from being purged, users should fetch output
that they wish to retain and save it as a regular data set.   In
WYIBUR, save the output in PRINT or EDIT format.  In ISO, use the
SAVE command of 2UEUE.

Large volumes of output should not be held in the queue.  Output
should either be printed immediately, written directly to a data
set for perusal, or written to a tape for printing later.

4.3.8    Print Trains and Special Forms

Provisions have been made to accommodate a variety of paper,
ribbon, and print train requirements for printed output.  The
user may establish a special form other than those listed below
or may request a combination not listed by contacting the shift
supervisor at the NCC or WIC (see Appendix A for telephone
number). Once the required form number is Known, the form may be
requested in one of three ways =

     a.  The form number xxxx may be specified in the JOB
         card by using the forms subparameter of the
         accounting information parameter.  All printed
         output for the job will be printed on the specified

     b.  The form number may be specified on individual DD
         statements, as follows:

          //ddname DD SYSOUT=(J,,xxxx)

     c.  The FORMS keyword parameter of the /*JOBPARM
         statement may be used (see Appendix C).

Certain terms will be used extensively:

    Print train:  A cartridge used on certain printers to change
    the characters which may be printed.  Those available to  the
    NCC are:

    PN:  Uppercase letters and numbers, with the following
    special characters:

      Character        Hexadecimal             Description

                          X'UB'                  period
           <              X'MC'                  less than
           (              X'MD1                  left paren
           +              X'ME1                  plus
           I              X'MF1                  vertical bar
           £              X'SO1                  ampersand

                                                         JAN 1988
    TN:   All the characters in the PN train plus the
    following special characters =
        abcdef ghi
                                              cent sign
                                              exclamation point
                                              lower case letters
                                              left brace
                                              less or equal
                                              superscript left paren
                                              superscript plus
                                              lower case letters
                                              right brace
                                              superscript right paren
                                              filled box
                                              degree(s )
                                              lower case letters
                                              left lower corner
                                              left upper corner
                                              left bracket
                                              greater or equal
                                              filled dot
                                              superscript numbers
                                              right lower corner
                                              right upper corner
                                              right bracket
                                              not equal

Terminal keyboards have the basic TN chain set, but there are a
number of differences in the special characters available.

    Ribbon:   The nylon medium that carries the ink.  Hammers
    impress  the paper on the ribbon to produce character images.
    Since nylon ribbons are used many times, the ink density
    varies considerably between a new ribbon and one about to be
    replaced.   A user who desires document quality print
    should specify the IBM 3800 printer at NCC or the Datagraphix
    9835 printer at MIC Production Services.

    Line density:   Refers to the number of lines printed per
    vertical inch.  Two line densities are available at the NCC:
    six and  eight lines per inch.  Six lines per inch is standard
    for most applications.





1 91
1 - X'89'
i _ x'99'
'A1 '
1 - X1 A9'
' AB'
1 - X'B9'

                            JAN 1988
Available forms are divided into three categories:  various
combinations o± print train, lines per inch, and plain lined or
unlined paper; various special forms for which a print train and
line density have already been established; and various
configurations of self-stick labels for which print train and
line density have also been established.

Plain Lined or Unlined Paper

Output to be printed with the PN train on single-part unlined
14-7/8 by 11 inch paper at six lines per inch is the standard.
No special form number need be given for such output. Other vari-
ations of plain-paper output may be specified by a four-digit
form number.  Forms requiring lined paper in multipart or sizes
other than 14-7/8 by 11 must be ordered specially. The Computer
Operations shift supervisor at NCC or WIC should be contacted two
weeks before usage in order to ensure that sufficient stock is on
hand (see Appendix A for telephone number).

The form number, which may be specified for both NCC and WIC
sites, is defined as follows:

     First digit:  Specifies the line density as either 6 or 8
     lines per vertical inch of paper

     Second digit:  Controls the print train.

            Value              Train
     Third digit:  Specifies the paper desired as follows:

            Value              Paper Description

14-7/8 by 11,  unlined
14-7/8 by 1 1 ,  lined
 9-1/2 by 11,  unlined
 9-1/2 by 11,  unlined, punched for
 three-ring  binder
12 x 8-1/2 unlined 3800 only;
reduced print
     Fourth digit:  Specifies the number of parts for the paper as
     a value from one to six.

NCC-IBM USER'S GUIDE                                     JAN  1988

The default form number is Form 6111.  It can be seen from the
table above that this would provide;

     o    6 lines per inch

     o    PN Train

     o    14-7/8 x 11, unlined paper

     o    One-part paper

Preprinted Forms

The following form numbers have been assigned to the indicated
combinations of preprinted paper, print train, and ribbon.  Each
of these forms will be printed at six lines per inch unless
otherwise noted. Additional form numbers may be established by
contacting the Computer Operations shift supervisor at NCC or
MIC.   The preprinted form numbers are not available at the WIG.
Any preprinted or special forms must be submibted to the NCC  or
MIC shift supervisor.

Number      Train     Description of Form (size in inches)

2401           PN      EPA Estimate of Municipal Wastewater
                      Treatment Facility Requirements.  Size
                      is 14-7/8 by 11.

2409           PN      National Pollution Discharge Elimination Systems

Self-Stick Labels

A variety of configurations of self-stick labels is available.
In all cases the labels are rectangular with slightly rounded,
1/8-inch radius corners.  Unless otherwise noted, all labels  will
be printed using the PN train at six lines per inch.  The
following chart associates each valid label form number with  the
description and dimensions of the labels it specifies.   The
dimensions referred to are all in inches and are =

A - the vertical dimension between successive occurrences
    of channel 1 on the carriage-control tape

B - the vertical dimension of the label

C - the horizontal dimension of the label

D - the vertical dimension between successive labels

E - the horizontal dimension between successive labels

                                                JAN 1988
No. A B
-1/2 1-7/1
-1/2 1-7/1
2306 1-1/2 1-7/1
1-1/2 3. 1
1 1/8
1 1/8
1-1/2 4-1/8
of Labels
1 -up
1 -up
                 Figure 4-5.  Self-Stick Labels

Note that self-stick labels cannot be used with the IBM 3800
Bursting and Decollating
Multipart printed output at the NCC will ^e decollated (divided
into separate segments, one for each copy with the carbon
removed) without any intervention from the user.  If decollation
is not desired, the user should contact Data Processing Support
Services before the job is submitted.  (See Appendix A for
telephone number.)

Multipart paper cannot be used with the 3800 laser printer at
NCC.  However, bursting of output as well as trimming of the
sprocket feed holes along the edge can be requested by specifying
BURST=Y on the SYSOUT card for the output; (e.g.,
SYSOUT=F,BURST=Y) (see Section 4.3.1).
 Plotter Output
The National Computer Center provides CalComp 1051 plotter
service at both the RTP and Washington Information Center
facilities.  Users local to either site may expect 24-hour
turnaround for their plotter output.  Mailing of plotter output
to the Regions is also performed at both facilities .

An assortment of graphics packages and subroutine libraries
capable of producing graphics on plotters as well as  graphics
terminals and line printers is available to NCC-IBM users.
Graphics software information is contained in Section 7.4.
Graphics hardware information is contained in Section

NCC-IBM USER'S GUIDE                                     JAN  1988



The main sources of information on the ISO functions are
available in the following:

     a.  IBM ISO manuals such as ISO Command Language
         Reference Manual  (Order No. GC28-0646) and ISO
         Command Language  Reference (Order No. SC28-1134).

     b.  The HELP command  available to the user during his
         terminal session.

Once the user has logged onto TSO, he may issue the HELP command.
If it is given with no operands, he will be provided with a list
of commands currently available on the system.

For information on any particular command, the user should enter
the following:

     HELP command-name

The following will appear:

     a.  A brief description of the command's function.

     b.  Format and syntax.

     c.  A description of  each available operand and
         defaults, if any.               '

If the user is currently under a particular command, he may enter
HELP with a subcommand name for similar details on its use.   For
example, if a user is editing under BED (alias E or EDIT), he
might enter HELP CHANGE.

To obtain the full options of the HELP command, the user should
enter HELP HELP.

TSO commands may be issued through BATCH (background) by using
the following JCL:

     //   EXEC TSOBATCH[,PRINT=class]
     //SYSIN DD *
     (TSO commands)

NCC-IBM USER'S GUIDE                                     JAN  1988

5.1.1    PRINTOFF

Some utilities are available under TSO.  The PRINTOFF function is
discussed below as well as being documented in HELP.

The PRINTOFF command can be used to obtain printed copy of the
data sets through foreground copying to SYSOUT.  The command
prints a data set(s) which is sequential or partitioned, blocked
or unblocked, with fixed or variable record length less than  or
equal to 256 (255 without carriage control).

The calling sequence is as follows:

     PRINTOFF (dslist) bin CLASS(class) DEST(remote)
         CAPS/ASIS [UCS(image) HTEXT 1
         [ASA ]
         [VOLUME (volser ) ]

     Required - dslist, bin


                ASIS, HEADING, DEST(RMTO)

where =

dslist (data set list) contains the names of data sets to be
printed or punched.   The names should follow TSO convention,  in
quotes if the primary index differs from the User-ID, or without
quotes and with the primary index omitted if it is the same as
the User-ID.

bin is the required four-character output bin number.  To route
to different box numbers for two individual PRINTOFF commands,
where the first has not yet completed printing, LOGOFF and LOGON
before issuing the second PRINTOFF command.

CLASS(class) is the SYSOUT class in which output is to be
printed.   Defaults to A.

DEST(remote) is the remote location to which SYSOUT data sets are

NCC-IBM USER'S GUIDE                                     JAN  1988

to be routed of form Rnnn, RMnnn, or RMTnnn.

COPIES(nnn) is the number of copies to be printed.  Defaults  to
1 .

OUTLIKoutlimit) is the output limit in number of lines  (see LINES
in Section 4.1.1).  May be 1-6 digits.

HOLD indicates that the output is to be placed on a HOLD queue
upon deallocation; NOHOLD is default.

LIST indicates that member names will be printed; using LIST
without the PRINT option will list just the member names in each
PDS in dslist.  NOLIST indicates that member names are not to be

PRINT indicates that members are printed (following member name
if LIST option given).   NOPRINT indicates that members are not to
be printed.

NOMSG suppresses the messages to the terminal.

CAPS indicates that the output is to be converted to upper case
prior to  printing.

ASIS indicates that output is printed without conversion to upper
case .

UCS(image) is the print image to be used when printed, e.g.,  TN
for upper and lower case; PN for all caps.

TEXT indicates that the output is considered to be text.  This
operand changes defaults to be ASIS UCS(TN).

FORMS(form) is the form to be used when the data set is printed
(see Section 4 . 3.7).

HEADING indicates that the output is to have heading information
generated by PRINTOFF;  header will contain data set name.
NOHEADING indicates that the output is not to have heading

SNUH indicates that the last eight columns of fixed-length
records or the first eight columns of variable length records
will not  be printed.   This allows sequence numbers to be
suppressed from the print.

ASA indicates that data set contains ASA carriage controls in
column 1.  Not necessary if RECFM indicates carriage control.

NCC-IBM USER'S GUIDE                                     JAN 1988

VOLUME(volser) indicates the volume on which data sets to be
printed are found.  This volume serial will be used for all data
sets specified in the data set list.  Not required for catalogued
data sets.

BURST indicates that the output is to be separated into sheets.
Otherwise,  output is printed in normal fanfold mode.
5.1.2    fiUEUE Command

The fiUEUE command provides ISO users access to JES2 data sets
residing on the SPOOL packs and displays information on current
status of JES2 processing queues (execution/print/punch).

A ISO user may examine any JES2 data set (JCL, JES2 JOBLOG, job
messages or any SYSIN or SYSOUT data set for any JOB submitted by
that TSO user).  fiUEUE may examine these data sets regardless of
HOLD status for the data set; that is, a user may browse a
SYSOUT=A data set even though it was not submitted with HOLD=YES
on the SYSOUT DD card.

fiUEUE also handles 'spin-off SYSOUT data sets from PRINTOFF
commands in TSO sessions.  The DD subcommand shows the spin
status and whether the data set has already been printed or not.
You may browse spin SYSOUT with this release of BUEUE.   The
command is as follows =

     BUEUE  cmdl oper1 ...

Required - none

Alias - fi

There are no standard operands to the fiUEUE command.  Any other
data entered on the same line as the fiUEUE command will be taken
as a subcommand for fiUEUE to process .

(Note:  in all the commands shown below, the lower case letters
        in the command may be omitted.  So the LOCATE command
        may be either 'LO'  or 'LOcate'.  The FETCH command
        may be either 'FEtch' or 'FE'.)

                                    JAN 1988
System Displays
   [B/S/T ]
                     LOcate (level)
Input Queue


Output fiueue



   [class Hdest ]
   [class Hdest ]
   [class Hdest ]
                     DO [class Hdest ]
                     AO [class Hdest ]
                     HO [class ] [dest]

Jobs executing or on
RDR/PRT/PUN devices.
Display TSO users.
Display started tasks.
Elsplay CPU Batch/STC/TSO
Job status.   Default for
status is * (TSO ID).

Display input queues by
Display all input jobs.
Display available jobs.
Display held jobs.

Display output queues by
class and record totals.

Display all output jobs.
Display available output.
Display held output.
NOTE:   Class may be A-Z, 0-9 for specific class, or * for
       any class.  Dest may be LOCAL or R1 to R255 for
       specific remote, blank implies any destination.
Job Modification
CAncel jobid
                   Cancel job from input or
                   or execution.  Delete
                   output if purge is
                   specified .
                   Functions like TSO
                   command!  'CANCEL
                   jobname  [ PURGE ] ' .
                     ROute job* rmt#
                   Route job to given
                   remote or "local".
                   If no rmt# given,
                   local is default.
                   Alias names: "RJ",
                   "PJ", and "PRINT".
                     REqueue jobid class
                                        Change held (Class
                                        P) SYSOUT class to
                                        new class.
                                        Functions like TSO
                                        command  'OUTPUT

NCC-IBM USER'S GUIDE                                      JAN  1988

                     DElete jobid       Delete held  (Class
                                        P) output.
                                        Functions  like  ISO
                                        command  'OUTPUT
                                        jobname  DELETE1.

NOTE:  Only JOBS submitted under your LOGON ID may be
       processed by these four commands.  REqueue  and DElete
       commands only operate on HELD SYSOUT data sets.

                                          JAN  1988

     H/HElp cmd/msgid

     STatus jobid
     JCL jobid [R]

Help for a given
command or error
message ID.  Message
ID's start with a
'B' followed by a
four- digit number.
Functions like TSO
command ' HELP' .

Display job.
List job JCL.  The R
option specifies
that 'raw'
(unconverted) JCL be
displayed (input JCL
as it was
List JES2 job log.
Only available after
job has completed
List job messages.
JES2 DD summary
List JES2 data set.
Obtain dsid values
by using the DD
     Find 'string' [ COL (:,-, ee ) ]
                        Find next occurrence
                        of 'string1 in the
                        data .
     FA11 'string' [COL(ss,ee)l
                        Find all occurrence
                        of 'string' in the
NOTE:  String must be in quotes; COL default
is all.
                     JLOG jobid
                     JMsg jobid
                     DD jobid

                     FEtch jobid dsid
                     DO/+  *
                     UP/-  *
                NOTE:  Default value to


                        Move forward # lines.
                        Move backward * lines.
                        * is 1 .

                        Max to top of data set
                        Top of current data
                        Bottom of current data

                                               JAN 1988

           SAve dsname
               Max to bottom of data set.
               Relative line number.

               L(S,E) where S is the starting
               line  number and E is the
               ending line number.   The
               default is the entire data
               D(NEW,OLD,MOD) Copy
               current data set to
               a ISO data set.
               Dsname is fully
               qualified data set
               name  if enclosed in
               quotes.  The default
               disposition is NEW.
               MOD copies the output
               to the end of a
               sequential data set.
               NEW creates a data set.
               OLD uses a previously
               defined data set.
NOTE:  'Jobid1 may either be the JOBNAME (e.g. XXX1) or
       the JES2 job number for the job (e.g., 136).
       •Jobname1 means only the JOBNAME may be specified
       'Job*' means only the JES2 job number may be
If job XXX1  is
subcommand may
JES2 job number
be either:
136,  then the DD
          DD XXXI or DD 136

 If the user fetches an entire job -- that is,
 the JCL plus all output data sets -- carriage
 control will be assumed throughout.   If the
 user wants to fetch output which does not have
 carriage control,  he must do so by fetching that
 output independent of the rest of the job.

 2UEUE is currently a stable product and no further
 enhancements will  be made.  Users with 3279 type
 terminals are encouraged to use the IBM SDSF product
 (refer to Section  5.1.M).

NCC-IBM USER'S GUIDE                                      JAN  1988

5.1.3    JES2 Global Status Tracking

JES2 global status tracking allows the user to determine  the
real-time status of a user's job at different nodes in the
network.  The ISO command 2X ships a JES2 request to the
specified node.  The remote JES2 builds display responses and
routes the responses asynchronously back to the ISO user.   The
remote JES2 will only send back status information of jobs
authorized for display by the local node.

The command is as follows:

     BX node prefix

Where node is the number indicating the NJE node where the  status
request will be routed.  If the node is omitted, the status
request is issued locally.   A table of NJE node numbers is  listed

Where prefix is the specified job name prefix.  Status
information will be displayed for all jobs whose job name begins
with the specified prefix.   If the prefix is omitted, the prefix
defaults to the User-ID.

The resulting display response contains the following
information =

     o  Time of response.

     o  Job number.

     o  Job name.

     o  Remote node name.

     o  Job status.

     o  Programmer name field.

A ISO help member is available by specifying HELP 2X.

5.1.4    Spool Display Support Facility

The Spool Display and Search Facility (SDSF) is a system
management aid which allows the user to analyze and control the
operation of an MVS/JES2-based system efficiently, providing
information about the system and its users on a realtime  basis.
The user can dynamically display the MVS system log and JES2
queues (input, output, ISO-held output, and active).  The user
has the ability to do the following ••

NCC-IBM USER'S GUIDE                                     JAN  1988

     a.  Scroll in any of four directions.

     b.  Search for occurrences of a given character string.

     c.  Print selected portions of the data.

     d.  Review job output in either OUTPUT or HEDD2UES.

     e.  Reroute jobs from one remote to another, cancel
         jobs in execution, change the job class, or purge
         job output.

For security reasons, some actions on a particular job are
limited to only your jobs.

SDSF operates on 3270 display stations as a menu option of ISPF
(option 8).

Refer to the TSO/ISPF User Reference Guide for more information
on the Spool Display Support Facility.

5.1.5    PC-IMF File Transfer

The file transfer software from IBM is called IBM 3270-PC File
Transfer Program (5665-311).  This software allows IBM PC's to
send and receive files from ISO on the NCC system.  The PC must
be attached  to a 3274/3276 controller through an IRMA board which
provides 3278 capabilities on the PC.

Complete documentation on the file transfer command is in the TSO
section of the DCA IRMA FT/3270 Reference Guide.

5.1.6    Job Status Tracking System

The Job Status Tracking System (JSTS) allows users to track the
activity of  a job as it progresses through the IBM network to get
a complete history of post-time job activity at different nodes.
At the origin node, the JSTS job automatically reads the instream
data to update the JSTS VSAM data base.  JSTS jobs are run and
executed periodically all day long to update the data base every
1 5 minutes.

With this information, the user can determine the time a
particular job was submitted, when it began and ended execution,
how many lines were printed, and the device and form on which the
output was printed.

A job's complete history is observed from a display which
contains the following:

NCC-IBM USER'S GUIDE                                     JAN 1988

     o  Job name

     o  Programmer name field

     o  Remote node name

     o  Job number

     o  Job status

     o  Total number of lines

     o  Date and time of each phase

     o  Job destination

     o  User-ID of the submitter

JSTS provides user-friendly panels for reviewing status records
and can be accessed one of two ways.   JSTS operates as a menu
option of ISPF (Option E,J) or,  if you are already in ISPF, enter
the TSO JSTS command.

Refer to the on-line JSTS User's Guide for more information on
the Job Status Tracking System (located in data set

5.1.7    Bulk Data Transfer

Bulk Data Transfer is an IBM licensed program that provides
facilities to transfer data sets from one computer to another.
BDT is available on the NCC IBM 3090  and at each regional Logical
Mainframe site thus allowing data sets to be moved from any one
of these computers to the other.

BDT is transaction driven.  Each request for a data set transfer
is submitted to the BDT program via a transaction.  Each
transaction describes the two computers involved in the transfer
and the data set to be copied.

Transaction requests to BDT can be executed in either of the
following ways =

     o  Interactively by executing BDT ISPF full-screen dialogs.

     o  In batch by executing program BDTBATCH.    Interactive Execution
                              5-1 1

NCC-IBM USER'S GUIDE                                     JAN  1988

BDT ISPF full-screen, interactive dialogs for building,
executing, and storing BDT transactions are available from the
EPA/NCC Options Menu.  Either Option BDT or B will provide BDT
transaction facilities.  Option BDT presents "vanilla" dialogs.
Option B presents NCC customized dialogs.    Batch Execution

The alternative to interactive execution of BDT transactions  is
shown in the following example.  The job (JCL) and the
transaction (SYSIN) are first created in an Operating System  data
set using an editor and then submitted to the operating system as
a batch job.

     //jobname JOB...
     //stepname EXEC PGM=BDTBATCH
     //SYSIN    DD  *
     Q JOBNAME (jobname) PROGRAMMERCname)
     FROM DATASETCfrom-dataset-name) MSG(NOTIFY=user-id)
        LOCCfrom-node) SHR DAP(PDS) CSOPT(DUP) PARMS(R=Y)
     TO DATASET(to-dataset-name)
        SHR DAP(PDS)  LOC(to-node) UNIT(DISK)
        VOLUMECvolser )
     /EOT    BDT Features

The following are some BDT features:

Transaction fiueue Integrity (TBI).  TBI ensures that each file-
to-file transaction reaches the BDT work queue and successfully

Direct Data Transfer.   BDT transmits data directly to the
receiving  data set without copying it to intermediate storage.

Sequential and Partitioned Data Set Support.  BDT will transmit
sequential data sets,  partitioned data sets, or selected members
of a partitioned data set.  Transmission can be sequential-to-
sequential, partitioned-to-partitioned,  sequential-to-
partitioned,  or partitioned-to-sequential.

DASD and Tape Support.  BDT will transfer data from DASD-to-DASD,
tape-to-tape, DASD-to-tape, or tape-to-DASD.

NCC-IBM USER'S GUIDE                                     JAN 1988

Mixed Data Set Attributes.   BDT allows reformatting of data sets
by permitting differing record formats, record lengths, and block
sizes betueen the TO and FROM data sets.

Intra-node Copy.  BDT can be used to copy data from one data set
to another when both data sets reside on the same computer.

Dynamic Allocation of Data Sets.   BDT will dynamically allocate a
data set receiving the data if this data set does not already

RACF Support.  BDT provides RACF parameters to protect data sets.

Checkpointing.  BDT checkpoints the progress of data transfers
and will reinitiate transfers from a checkpoint in the event of
system failure.  This feature eliminates the retransmission of a
large data set following a system failure.

Generic Master Job Description Support.  Frequently executed
transactions can be stored in a Generic Master Job Description
(GMJD)  library.  BDT will execute the transaction by passing only
the member name of the GMJD library containing the transaction.
Prior to execution, overriding parameters are merged with the
GMJD transaction.

Messages.   BDT issues messages on a user basis which provide
status  information on individual transactions.  These messages
are delivered to the TSO session interactively submitting the
transaction.  For batch execution, the keyword/parameter
"MSG(NOTIFY=uid)" will cause the messages to be sent to TSO user

Commands to Check Progress.  BDT provides a variety of commands
that allow querying of data transfer transactions.    Documentation

On-line documentation is available from within a BDT ISPF full-
screen session by selecting option BDT from the EPA/NCC Options
Menu and entering HELP on the ISPF command line.  The information
presented is sufficient for understanding the interactive
creation, execution, and storage of BDT transactions.

The following documentation is available from IBM=

     MVS/Bulk Data Transfer, Version 2, General Information

     MVS/Bulk Data Transfer, Version 2, File to File Transaction
     Guide (SC23-0225-0)

NCC-IBM USER'S GUIDE                                     JAN  1988

     MVS/Bulk Data Transfer, Version 2, Commands  (SC23-0226-0)

     MVS/Bulk Data Transfer, Version 2, Messages  and Codes


IBM-supplied utilities are an integral part of the IBM system and
provide a variety of useful functions.  The following is a list
of the most-used subset of these utilities =

     IEFBR14     Allocates, deletes, and catalogs data sets

     IEHPROGM    Deletes, renames, catalogs, and  uncatalogs data sets

     IEBGENER    Copies sequential data sets

     IEBCOPY     Copies and/or compresses partitioned data sets

     IEHMOVE     Copies/moves sequential, partitioned, or direct
                 access data sets

     IEBISAM     Copies or lists ISAM data sets

     IEHLIST     Lists DASD volume information

     IDCAMS      Allocates and maintains VSAM data sets

The Linkage Editor is also an integral part of the IBM system; it
is discussed in Section 5.2.9.

Sample Job Control Language used with the utilities listed above
is as follows:

     //stepname  EXEC PGM=utility-name
     //SYSPRINT  DD SYSOUT=class
     //SYSUT1    DD DSN=input-data-set-name
     //SYSUT2    DD DSN=output-data-set-name
     //SYSIN     DD *
       (control statements)

Note: SYSPRINT prints utility messages with LRECL=121.

SYSUT1 and SYSUT2 may not be necessary; -!h.air use depends on the
utility and the desired function.

Control statements must be coded in columns 2 through 71.  To
continue a control statement, code a non-blank character in
column 72 and start the continuation in colum 16  of the next
line .

NCC-IBM USER'S GUIDE                                     JAN  1988

Full documentation is in "OS/VS2 MVS Utilities", manual number
GC26-3902.  A brief description of their use follows.
5.2.1    IEFBR14

IEFBR14 is not a utility in the true sense and is not covered in
the IBM Utilities Manual.  The program terminates immediately
with a return code of zero.  IEFBR14 allows system allocation
routines to be invoked so that the user can check JCL syntax and
allocate,  delete, catalog, and uncatalog data sets through JCL
alone.  It is also helpful in ensuring that enough space is
available  for data sets without invoking the main program.  An
example follows =

   //stepname  EXEC PGM=IEFBR14
   //ddnamel   DD DSN=new-data-set-name,DISP=(NEW,CATLG),
   //   SPACE=(6160,(30,6)),UNIT=DISK
   //ddname2   DD DSN=tape-data-set-name,DISP=(OLD,CATLG),
   //   UNIT=(TAPE62,,DEFER),DCB=DEN=4,
   //   VOL=SER=tape-volume

Although the tape volume data in this example will be catalogued,
the tape itself will  not be mounted if the user specifies DEFER
in the UNIT parameter.  IEFBR14 opens no data sets referenced in
the JCL.

IEFBR14 is also used  quite frequently as a preliminary step to
guarantee  that a particular data set allocated later in  the job
does not exist before that step.  An example follows:

   //stepname  EXEC PGM=IEFBR14
   //ddname    DD DSN=data-set-name,DISP=(OLD,DELETE)

Warning:  IEFBR14 should NOT be used to delete a member  of a
partitioned data set.  If this is attempted, the entire  data set
is deleted.
5.2.2    IEHPROGM

The functions of IEHPROGM are as follows:

     o   Scratches data sets or members of data sets on DASD

     o   Renames data sets or members of data sets on DASD

     o   Catalogs or uncatalogs data sets.

NCC-IBM USER'S GUIDE                                     JAN  1988

IEHPROGM is a data set maintenance utility.  Usually the user can
use the disposition field (DISP) in his JCL for cataloging or
uncataloging, but if a user wishes to scratch a data set without
deleting a catalog entry, he must use IEHPROGM.  Functions under
TSO such as RENAME and SCRATCH may be used, but these affect  the
catalog entry for the data sets.

The basic JCL format is as follows:

     //ddname   DD VOL=SER=volume,DISP=OLD,UNIT=SYSDA
     //SYSIN    DD *
       (control statements)

The DD statement pointing to SYS 1.SVCLIB need only be used when a
reference to the system catalog will be made; i.e., CATLG and
UNCATLG.  One of the major drawbacks in the use of IEHPROGM is
that the user must know what explicit volumes are being occupied
by data sets referenced in the control statements.  TSO commands
do not require this information.

The control statements relevant to general use are as follows.
The symbols [] indicate optional information.

     a.  SCRATCH scratches a data set or a member from a
         direct access volume.

         [ , PURGE]

     b.  RENAME changes the name or alias of a data set or
         member residing on a direct access volume.


     c.  CATLG generates an entry in the index of a catalog.


     d.  UNCATLG removes an entry from the lowest level
         index of the catalog.


NCC-IBM USER'S GUIDE                                     JAN  1988

In all examples VOL=device=list provides device type  (e.g., 3350)
and the volume(s) on which the data set resides.  These volumes
must be allocated by DD statements in the JCL.

IEHPROGM can scratch any type of data set:  sequential,
partitioned, direct or indexed-sequential (except VSAM files  use
IDCAMS).  The operation will not affect any catalog entry  for
that particular name.  Care should be taken to ensure that the
data set is not being used concurrently by another program.

MEMBER=name will scratch only one member of the data  set.  When
using this option, it is wise to use the MEMBER operand
immediately after the SCRATCH keyword.  A simple error in  typing
(e.g., an inadvertent blank) can otherwise result in  losing an
entire data set.

PURGE will cause the data set to be scratched even though  the
expiration date has not expired.

RENAME will rename either an entire data set  or one member within
it.  (Warning:  RENAME will not update the catalog entry for  the
data set).

For both SCRATCH and RENAME, the user must have ALTER privileges
if the data set is RACF-protected.

With CATLG/UNCATLG an entry will be made in or deleted from the
appropriate catalog.  A DD card referencing SYS1.SVCLIB must  be
5.2.3    IDCAMS

The VSAM utility IDCAMS (Access Method Services) provides data
set maintenance functions without the need to write user program,
including the following:

    o  Defining VSAM data sets
    o  Deleting VSAM data sets
    o  Data set recovery and backup
    o  Data set printing
    o  Creating alternate indexes for KSDS or ESDS
    o  Catalog listings
    o  Altering tuning parameters
    o  Copying ISAM, VSAM or SAM data sets to VSAM or SAM
       data sets

IDCAMS also replaces the function of IEHIIST which previously did
a LIST catalog.

For example, to list a catalog for a list of data sets with a

NCC-IBM USER'S GUIDE                                     JAN  1988

specific User-ID/account combination, the following JCL may be
used :

     // . . . JOB. . .
     //SYSIN    DD *
        LISTCAT  L(iiiaaaa)

For a more detailed use of IDCAMS, the data set JUSD.DFEF.DATA
contains information and sample JCL showing how to use the IDCAMS
utility.   Other documentation is available from IBM;

     a.  MVS/Extended Architecture Integrated Catalog
         Administration:  Access Method Services Reference
         (GC26-4019-1) contains information about the access
         method services command used to manipulate ICF
         catalogs and VSAM data sets.

     b.  MVS/Extended Architecture VSAM Administration Guide
         (GC26-MO 1 5-1 ) describes the use of virtual storage
         access method services commands, VSAM macro
         instructions, and JCI to process VSAM data sets.

     c.  MVS/Extended Architecture Catalog Administration
         Guide (GC26-M041) describes how to use the
         integrated catalog facility.

If a generation data group is to be created, the first operation
must be to define the generation data group.  IDCAMS replaces the
function of IEHPROGM which previously built generation data
groups.  A generation data group is a group of data sets having
identical high level qualifiers.  The last qualifier,  the
qualified name,  is of the form .GxxxxVyy where xxxx is the
generation number and yy is the version number.   Each time the
user creates a new generation data set (the "+1" generation), the
data set name is automatically created with a generation number
one greater than the highest current data set generation.  The
user can refer to his data sets using the common high level
indices plus the relative generation.

For example, a user wishes to establish a Generation Data Group
(GDG)  for all users in account aaaa.   He never wants more than
three  data sets catalogued.  When a fourth data set is created,
he wants the oldest one of the three existing data sets to be
uncatalogued automatically and the data set to be deleted
(scratched).  First, he defines the index using the following

NCC-IBM USER'S GUIDE                                     JAN  1988

     //. . .JOB. . .
     //SYSIN     DD    *
       DEFINE GDG ( NAME ( uuuaaaa . name ) -
                   SCRATCH -
                   LIMIT(nn) )
After creating the index, a subsequent job can create and catalog
a new data set in the generation data group with the following DD
statement using the model DSCB designated by EPA:

     /xddname     DD DSN=aaaa . name (+ 1 ), DISP= ( NEW , CATLG ),...
     //              DCB=(MODEL. GDG. DSCB, RECFM= .  . . ,LRCL=. . .

This DD statement may include any additional parameters needed
for the new data set, such as UNIT,  and SPACE for disk data sets
or LABEL for tape data sets .   You must specify a DCB parameter
and specify the model DSCB as MODEL . GDG . DSCB .

After creating and cataloging three or more data sets in the
generation data group, a user may refer to the "oldest" -- that
is, the data set with the lowest generation number — with a DD
statement similar to the following:

     //ddname     DD=DSN=aaaa. name (-2) ,DISP=SHR

Notice that the oldest generation is specified as the number of
entries (3 in this example) minus one.  The "current" data set is
aaaa . name ( 0 ) .   In this manner, 'the user is not required to change
JCL to refer to changing generations of data sets even though
their actual names differ.

The IDCAMS delete command removes the description of the
generation data group from the system.  All entries for this GDG
must be deleted before issuing the command to delete the
generation data group base catalog entry.  Refer to the MVS/XA
Integrated Catalog:  Access Method Services Reference Manual for
more details .
5.2.4    IEBGENER

The functions of IEBGENER are as follows:

     o  Copies sequential data sets.

     o  Copies individual members of partitioned data sets

     o  Reblocks data sets.


NCC-IBM USER'S GUIDE                                      JAN  1988

Although IEBGENER has other functions to rearrange and replace
fields within records, its most common uses are those listed
above.  The user may refer to the IBM manual MVS/Extended
Architecture Data Administration Utilities  (GC26-4018-1)  for  the
more esoteric functions.

The basic JCL format is as follows:

     //SYSPRINT  DD SYSOUT=class
     //SYSUT1    DD DSN=input-data-set,DISP=SHR
     //SYSUT2    DD DSN=output-data-set,DISP=SHR
     //SYSIN     DD DUMMY

If the output is to be reblocked, the data  sets still must have
the same RECFM and IRECL attributes.  Only  the BLKSIZE may be
changed.  To copy a single member,  the user should indicate that
member on the DD statement.  The system always treats this as a
sequential data set:

     //SYSUT1 DD DSN=iiiaaaa.PDSCmember),DISP=SHR

Often IEBGENER is used to print a particular data set by
specifying SYSOUT=class for SYSUT2.  SYSPRINT may be set  to DUMMY
if desired.  IEBGENER is also used  to save  a card deck on disk:

     //SYSUT1  DD *
        (card deck)
     //           UNIT=DISK,SPACE=(qty,(pri,sec),RLSE)

5.2.5    IEBCOPY

The functions of IEBCOPY are as follows =

     o   Copies a partitioned data  set.  Often used to
         expand allocation or change directory space.

     o   Creates a backup copy on tape of a PDS.

     o   Compresses a partitioned data set  in place.

IEBCOPY has a subset of functions to those  of IEHMOVE but is
easier to use and more efficient in performing those common

The basic JCL format is as follows:

NCC-IBM USER'S GUIDE                                     JAN  1988

     //in-dd    DD DSN=input-data-set,DISP=SHR
     //out-dd   DD DSN=output-data-set,DISP=...
     //SYSIN    DD *
      COPY OUTDD=out-dd,INDD=in-dd

For all but a compress-in-place,  the output  data set can be
reblocked.  In the basic JCL given, if in-dd=SYSUT1 and out-
dd=SYSUT2, SYSIN can be DUMMY; IEBCOPY will  automatically assume
the control card;


The full syntax of all the control statements available to
IEBCOPY is shown below.  (All keywords may be abbreviated to  the
first letter; e.g., C I=in-dd,0=out-dd.)

     a.  COPY - indicates the beginning of a COPY operation.



                ddname1[ ,ddname2  ] [ ,(ddname3,R). . . ]


                ((ddname1,R)[ ,ddname2 ] . . . )


NCC-IBM USER'S GUIDE                                     JAN  1988

     b.  SELECT specifies which members in the input data set are
         to be copied.

         MEMBER=(name 1,name2, . . . )


                ((name 1,neunarae,R),...)


                ((name 1,neuname),...)


                ((name 1,,R), . . . )

     c.  EXCLUDE specifies members in the input data set to be
         excluded from the copy step.

         MEMBER=(membername 1,membername2, . . .)

IEBCOPY can be used to  merge many data sets by specifying
multiple DD names in the INDD operand.  These will be searched in
the order specified by  the control card.  Only the first member
of a given name uill be copied unless the replace option
(ddname,R) is specified.   The LIST=NO operand will prohibit the
listing of the copied members on the SYSPRINT file.

SELECT and EXCLUDE are  mutually exclusive operands which limit
the scope of the COPY operation.   SELECT will choose only those
members specified and may be used to assign a new name during the
copy.  The R option is  also available to indicate that
identically named members in the output data set will be

EXCLUDE will prohibit copying of the explicitly named members.

Any number of SELECT or EXCLUDE statements may follow a COPY
operation.  Since many  COPY operations can be initiated in the
same run,  the SELECT or EXCLUDE cards concern only the COPY
statement immediately preceding them.

To compress a data set  in place,  the user should use the same
DDNAME reference for both input and output ••

     COPY INDD=ddname,OUTDD=ddname

NCC-IBM USER'S GUIDE                                     JAN  1988

The example below copies the partitioned data set referenced  by
the DD statement INPDS and creates the data set referenced by the
DD statement OUTPDS.  At the end of this job there will be two
copies of the data set.

  //. . .       JOB
  //INPDS    DD   DISP=SHR,DSNAME=iiiaaaa.oldfile
  //         DSNAME=iiiaaaa.newfile
  //         SPACE=(6160,(30,6,5)),UNIT=DISK
  //SYSIN    DD   *
5.2.6    IEHMOVE

The functions of IEHMOVE are as follows:

     o   Copies or moves sequential, partitioned, or direct
         data sets.  COPY leaves the original; MOVE
         scratches the original after the copy, and the
         catalog entry is changed.

     o   Automatically allocates space or it may be

     o   Reblocks the target data sets if requested.

Although IEHMOVE has many duplicated functions in IEBGENER and
IEBCOPY, it is more powerful because space allocation is done
automatically., and it can handle direct as well as partitioned
and sequential data sets.

The basic JCL format is as follows =

     //SYSPRINT  DD SYSOUT=class
     //ddname    DD VOL=REF=SYS1.SVCLIB,DISP-OLD
     //tape      DD DSN=dsname,DISP=...
     //disk      DD VOL=SER=volume,UNIT=SYSDR,DISP=OLD
     //SYSIN     DD *
        (control statements)

IEHMOVE must have one DD statement pointing to a permanently
mounted volume.  A reference to SYS 1.SVCLIB's volume is always
safe since it points to the system residence device.

NCC-IBM USER'S GUIDE                                     JAN  1988

If a data set is copied to or from a tape volume, a DD statement
must define it.   DD statements must also be provided for any
volumes referenced in the control statements.  To reference a
volume containing a particular catalogued data set, the DD
statement may be of the form below ••

     //ddname  DD VOL=REF=dsname,DISP=OLD

To change blocking, the user should preallocate the space with
the new DCB parameters.   COPY/MOVE DSNAME works on sequential
data sets; COPY/MOVE PDS, on partitioned data sets.  If the
target data set is tape, the original data set is unloaded; i.e.,
converted to 80-byte records and saved sequentially.  Using
IEHMOVE to move back from tape will load the data set.  It must
go back to the same device type as the original.  IEHMOVE will
automatically assign the same DCB attributes and space as was in
the original if space is not preallocated.

The COPY and MOVE statements have identical operands except for
the CATLG operand which is used only with the COPY statement.
The EXPAND operand is only relevant for partitioned data sets.

In the following examples information in I ] is optional.

     a.  COPY (or MOVE)  DSNAME copies (or moves) sequential
         or direct data sets.

          [ , FROM = device = list]
          [ ,CATLG ]
          [ ,RENAME = name ]
          t,FROMDD = ddname ]
          [ ,TODD = ddname ]
          [ ,UNLOAD ]

     b.  COPY (or MOVE)  PDS copies (or moves) a partitioned
         data set.

          [ ,FROM = device = serial ]
          [ ,EXPAND = nn ]
          [ ,UHCATLG ]
          [ ,CATLG ]
          [ ,RENAME = new-name ]
          [ ,FROMDD = ddname 1
          [ ,TODD = ddname ]
          [ ,UNLOAD ]

NCC-IBM USER'S GUIDE                                      JAN  1988

The keyword DSNAME may not be abbreviated as in OCL with  DSK.

FROM defines device type (e.g., 3350) and volume serial of  the
original data set.  It need not be given for a catalogued data

TO defines the device type and volume serial number of the  target
data set. It is required.

FROMDD=ddname is coded for tape input volumes where the ddname  is
that defining the tape.

TODD=ddname is coded for tape output volumes.  The DD statement
allocates a data set from which LABEL and DCB information is

UNCATLG uncatalogs the original data sets.  It is ignored if  the
FROM option is coded.

CATLG catalogs the new data set (used only with COPY).

RENAME=new-name renames the target data set.

EXPAND=n expands the directory of a PDS by n 256-byte blocks.
The user may also preallocate to accomplish this, in which  case
the EXPAND operand is ignored.  This operand is used only with
MOVE/COPY PDS statements.

Following the MOVE/COPY PDS statement the user may add INCLUDE,
EXCLUDE, SELECT, and REPLACE statements.  Several of any  one  may
be coded after a MOVE/COPY.  These cannot, i>e used for a data  set
being unloaded or loaded.

     INCLUDE  DSNAME=dsname

INCLUDE copies selected members from another PDS.  A DD statement
must be included pointing to the volume containing the data set.
Members of this data set will not be scratched even for a MOVE

If MEMBER .   omitted, all members are included.

FROM is \      nly for uncatalogued data sets.

     EXCLU,      'BER=membname

EXCLUDE rest.  ~   the scope of MOVE/COPY.  Many EXCLUDE
statements ma.     -pecified.  On a MOVE operation, these members
will be lost s  ce '-he entire data set is scratched.

NCC-IBM USER'S GUIDE                                     JAN  1988

     SELECT MEMBER=(name[,name]...)
            MEMBER=((name,newname) [ ,name ] . . . )

SELECT also restricts the scope in choosing  only those members
specified.  SELECT cannot be used with either EXCLUDE or REPLACE
to modify the same MOVE/COPY.  Members not  selected are lost  in a
MOVE operation since the  data set is destroyed .

Any selected member may optionally be renamed as it is moved  or

     REPLACE DSNAME=dsname

REPLACE excludes a member from the MOVE/COPY and replaces it  with
a member from another data set.  The new member must have the
same name as the old.
5.2.7    IEBISAM

The functions of IEBISAM are as follows =

     o   Copies an indexed sequential (ISAM) data set from
         one DASD volume to another.

     o   Creates a backup copy of an ISAM data set by
         unloading it into a sequential data set on a DASD
         or tape volume.

     o   Restores from  such a backup.

     o   Prints an ISAM data set.

     o   Reorganizes an ISAM data set to delete records
         previously marked for deletion.

The basic JCL format is as follows:

     //   EXEC  PGM=IEBISAM,PARM=fparm-irifo1
     //SYSUT1   DD define-input-data-set
     //SYSUT2   DD define-output-data-set

No utility control cards are required.  Necessary information
defining the desired operation is passed in the parameter field
of the EXEC statement.

The following options may be coded in the FARM field:


Parameter      SYSUT1

COPY            ISAM

UNLOAD          ISAM




                                JAN 1988

SYSUT2         Operation

 ISAM      Copies the ISAM data set

 SEP,       Converts ISAM to sequential
           (Default parameter value)

 ISAM      Loads an unloaded data set

 PRINT     ISAM data set printed in
           character format

 PRINT     Same as PRINTL except that
           output is in hexadecimal
To reorganize the data set, the user should submit a tuo-step
program:  the first step unloading the ISAM data set to tape; the
second, loading from that tape back to the original data set.

An example follows =

     // DCB=(RECFM=FB,LRECL=80,BLKSIZE=8000)
IEHLIST lists the entries in the directory (i.e., the members) of
a partitioned data set created by the Linkage Editor.

IEHLIST can list up to ten partitioned data set directories in a
single invocation.  A partitioned data set directory is made up
of blocks of 256 bytes, each containing one or more entries which
reflect member and/or alias names as well as attributes assigned
by the Linkage Editor.  From the listing the user can see what
members exist, their sizes, and various attributes.

The JCL format is as follows =

     //ddname   DD VOL=SER=dsvol,UNIT=DISK,DISP=OLD
     //SYSIN    DD *
       (control statements)

NCC-IBM USER'S GUIDE                                     JAN  1988

The control statement is as follows =

     IISTPDS DSNAME=dsname,VOL=device=volume,FORMAT

The dsname may be a list of up to ten names of partitioned data
sets.  If there are more than one,  the list is enclosed in

VOL=device=volume describes the type of device (e.g., 3350) and
the serial number of the volume on  which the data set(s) reside.
It must be allocated in the JCL.

FORMAT edits the listing into a readable report listing.

DUMP lists the output in unedited hexadecimal form.

If both DUMP and FORMAT are missing, an abbreviated edited format
is provided.

The basic functions of IEHLIST may  also be obtained from the  use
of the PDS command under TSO.  The  user may issue HELP PDS for
more information.

5.2.9    Linkage Editor

Each program starts as from one to  several separate source
modules.  These are assembled or compiled by one of the language
translators such as Assembler or COBOL.  The output of this
translation is an object module.   Before an object module can be
executed, it must be processed by the Linkage Editor which forms
a load module.   Each module in the  input to the Linkage Editor
may contain symbolic references to  control sections in other
modules; i.e.,  external references.  These symbols referred to by
an external reference must be either the name of a control
section or of an entry point in a control section.  The Linkage
Editor resolves these external references between control
sections and issues diagnostics when no match can be made.

The Linkage Editor used at NCC is compatible with MVS/370 and
MVS/XA operating systems.  This version is also compatible with
all existing JCL and supports Assembler H Ver 2 generated object

The main function of the Linkage Editor is to take a combination
of object modules, load modules,  and control statements and
combine these into a single output  load module.  Many other
functions are available but these will not be covered in this
discussion (refer to the IBM manuals OS/VS Linkage Editor and
Loader, GC26-3813 and OS/VS Message Library:  Linkage Editor  and

NCC-IBM USER'S GUIDE                                     JAN  1988

Loader Messages, GC 38-1007).  The Linkage Editor may also be
invoked in the TSO foreground with the LINK command.  The user
may issue HELP LINK under TSO for the syntax.

For more information the following manuals are also available
from IBM:

     a.  MVS/Extended Architecture:  Linkage Editor and
         Loader Messages (GC26-M020 ) .

     b.  MVS/Extended Architecture:  Linkage Editor and
         Loader (GC26-4011).

The specific DD names used to call the Linkage Editor are listed
below.  Additional DD statements may be used to allocate other
sources of input.   These may have any user-assigned names.

 DDNAME              Data Set                 Required?

SYSLIN            Primary input             yes

SYSLIB            Automatic call library    Only if automatic
                                            call mechanism
                                            is used

SYSUT1            Work data set             yes

SYSPRINT          Diagnostic messages       yes

SYSLMOD           Output module             yes    SYSLIN

The primary input can be both object modules and control
statements.  For example, the following assumes that previous
compiler steps have output object modules to temporary data set

     //       DD *
      NAME LOAD1

This primary input would form a load module of all the object
modules in ££OBJMOD.  This load module would be given the member
name LOAD1.  When executed, execution would start at the entry
point called CSECT1.

In most standard procedures with both compile and link steps,
this primary input will be given in the following form:

NCC-IBH USER'S GUIDE                                     JAN  1988

     //       DD DDNAME=SYSIN

The user may include his control statements by adding the DD
statement ••

     //LKED.SYSIN DD *
        (control statements)

This will have exactly the same effect as the concatenation shown
above.    SYSLIB

The automatic call mechanism is used to resolve external
references.  The call library defined on -'he SYSLIB DD statement
is searched.  This call library may contain either  (1) object
modules and control statements or (2) load modules.  It cannot
contain both.  Use of the NCAL option in the parameter field for
the Linkage Editor negates this mechanism.  The parameter LET
will allow the module to be marked as executable even if there
are unresolved external references.

SYSLIB may be a number of concatenated libraries.   They will be
searched in the order in which they are given.    SYSUT1

SYSUT1 is a sequential work data set used by the Linkage Editor.    SYSPRINT

The user may choose to put diagnostic messages from the Linkage
Editor into a data set rather than send them to the printer.  The
data set must be one of the following:


     DCB=(RECFM=FBA,LRECL=121,BLKSIZE=n*121) where  n is
     an integer not greater than MO.    SYSLMOD

SYSLMOD describes the output module  library which must be a
partitioned data set.  RECFM=U is assigned by the Linkage Editor.
A member name may be specified, but it is used only if a name was
not specified on a NAME control statement.

NCC-IBM USER'S GUIDE                                     JAN  1988    INCLUDE Control Statement

The INCLUDE control statement requests the Linkage Editor to  use
additional data sets as input.  These can be as follows:

     a.  Sequential data sets containing object modules
         and/or control statements.

     b.  Members of partitioned data sets containing object
         modules and/or control statements.

     c .  Load modules.

The INCLUDE statement specifies the DD name of a DD statement
describing the data set to be used as additional input.  The  DD
name may be any user-chosen name.  If the DD statement describes
a partitioned data set, the INCLUDE statement must contain the
name of each member to be used.

As an example, the following assumes that a load module NEWMOD is
to be formed from an object module in III.AAAA.OBJ and members
LOAD1 and LOAD2 in load library III.AAAA.LOAD.   The entry point
is at CSECT MAIN.  An old member NEWMOD is to be replaced.

   //  EXEC LKED
5.2.10    ICF Catalogs

The Integrated Catalog Facility (ICF) is a function of the IBM
Data Facility Extended Function (DFEF) program product.  ICF is <
functional replacement for VSAM Master Catalogs, VSAM User
Catalogs, and OS Control Volume (CVOL) catalogs with major
improvements in reliability, recovery, performance, and
usability.  All three types of catalogs can be combined into one
catalog structure doing away with VSAM ownership, etc.

ICF Catalogs do not support the LISTCAT function of IEHLIST or
the BLDG function of IEHPROGM.  They are handled by the DFEF
utility IDCAMS (see Section 5.2.3 for more information on

NCC-IBM USER'S GUIDE                                     JAN  1988

5.2.11    VSAM

The IBM Virtual Storage Access Method (VSAM) is designed as a
high performance replacement for ISAM.  Users are encouraged  to
convert from ISAM to VSAM.  ISAM support is being phased out  by
IBM.  New development in ISAM is discouraged.

The following data set organizations are supported by VSAM =

     a.  Key Sequenced Data Sets (KSDS) with fixed or
         variable length records in the data component, and
         an index component with sequential or keyed

     b.  Relative Record Data Sets (RRDS) with fixed length
         records and addressing by relative record number.

     c.  Entry Sequenced Data Sets (ESDS) with fixed or
         variable length records and addressing sequentially
         or by relative byte address.

VSAM maintains a comprehensive set of statistics to provide
information concerning utilization, performance, and access and
update counts.  It also provides device independence, program
independent tuning parameters, RACF security support, a range of
data sharing and integrity options, and a comprehensive set of
utilities.  NCC is using IBM's Data Facility Product (DFP) with
the Integrated Catalog Facility (see Section 5.2.10).

Documentation is listed in JUSD.VSAM.DATA(DOC ).

JCL considerations should be made when using VSAM.  Since VSAM
data sets are created and deleted by IDCAMS, the only JCL
parameters required to access them are DSN, and DISP of SHARE or

ISAM data sets can be converted to VSA11 KSDS by using the IDCAMS
DEFINE and REPRO functions.  The ISAM Interface Program (IPP)
provides an interface that allows ISAM file commands to be
translated to appropriate VSAM commands.  This permits ISAM
programs to access VSAM data sets without any changes.  The IPP
provides a valuable conversion aid until the ISAM program can be
receded.  A complete description of IPP :'..: found in the IBM VSAM
User's Guide available from IBM.

VSAMAID/XP provides the tools required to tune and monitor VSAM
clusters.  It is primarily for the performance analyst and the
capacity planning manager, but it can also be used for tuning
individual VSAM clusters.  Statistical history can be used to
produce cluster definition recommendations as follows =

NCC-IBM USER'S GUIDE                                     JAN 1988

     o  Attempt to reduce index levels .

     o  Maximize DASD utilization.

     o  Estimate required free space by projecting past growth.

     o  Estimate true allocation requirements based on--

        -  Calculated average record size.
           Device capacity for the selected CISIZE.
           Free space requirements.

For more information on the use of VSAMAID/XP, contact User
Support (see Appendix A for telephone number).


5.3.1    SYNCSORT

The NCC offers one software package, Syncsort Inc.'s SYNCSORT,
for sort/merge applications.   SYNCSORT offers all of the
capabilities included in IBM's Sort/Merge program as well as
certain extra features.  First, SYNCSORT allows the use of
fragmented disk space for sorting work areas.  This feature makes
it possible for a user to specify both a primary and secondary
space allocation rather than a single, larger, contiguous area;
the use of this technique significantly reduces the likelihood of
an ABEND during the sorting process.

Second, SYNCSORT automatically determines the size of the input
file and associated degree of randomness.   The program then uses
this information to determine the most efficient sorting
techniques to be used for that file.

The following JCL invokes the SYNCSORT procedure:

     //stepname EXEC SYNCSORT

Some of the enhancements incorporated in the latest version of

     o   Record reformatting includes new field manipulation
         capabilities for data conversion, editing,  and
         insertion of literals.

     o   Report formatting provides a full range of report
         formatting options.

     o   Multiple outputs that include the same or different

NCC-IBM USER'S GUIDE                                     JAN  1988

         data can be created from one sorted file.

Tape sorts are not available on the IBM system.  Users should use
MAXSORT if large volume sorts are to be performed.  Since tape
sorts are not used, a SORTLIB DD card is unnecessary.

Release 2.5 SYNSCORT Programmer's Guide may be purchased by
contacting =

                   Syncsort Incorporated
                   50 Tice Boulevard
                   Woodcliff Lake, NJ 07675

                   Phone (201) 930-9700

On-line documentation on MAXSORT is available in

5.3.2    KWIC/KVIOC

The Key-Word-In-Context (KWIC) system is available on the NCC-IBM
system.  KWIC, a group of COBOL programs that was originally
designed to handle bibliographical information, provides a simple
input procedure and flexible multipoint output of almost any  kind
of non-computational information.  The system, which can be used
for anything from a reprint file to a legal document system,
allows the automatic generation of various indexes based on data
provided by the user.  The indexes requested most frequently  are
those by author, title, Key-Word-In-Context, and Key-Word-Out-Of-

KWIC card image input format allows data input from punched cards
or via any of the standard system data editors such as fiED.    The
input may be in a variable field length string format (i.e.,  a
continuous string of characters separated by field identifiers)
or in fixed-field format.  Since preceding of the data is seldom
necessary, clerical personnel can normally perform data entry

The KWIC system requires that a unique identifier be assigned for
each item to be processed.  This may be a book call number,  a
sequential number, a date, or any meaningful combination of
alphanumeric characters.  The programs draw the data to be
processed from a group of records which are organized into not
more than 10 definable fields or categories of information.    A
group of records which relates to a particular unique identifier
is known as an information packet identifier, or number.

In the following example of KWIC input, and some possible
outputs,  missing data is denoted "...".

NCC-IBM USER'S GUIDE                                     JAN  1988




    ...AND HAZARDOUS  DUMPS                        UN-01
                      HAZARDOUS WASTES AND...      UN-01

The KWIC-formatted index always lines up the indexed words down
the middle of the report page with the unique number on the
extreme right of the page.  The index word is shown "in context"
of the surrounding words.  The ability is provided to exclude
non-significant words (STOPWORDS) from the index such as "and" in
the example.





The KWOC-formatted index always extracts the index word from  the
title and places it "out of context" in the left margin of the
page.   The titles containing that word follow, indented on a
separate line along with the associated unique number.  If an
index word occurs more than once in the title, the title is only
printed once.

These KWIC and KWOC examples show the basic report formats of the
KWIC system.   If authors or assigned index words had been used,
they would have been printed in the KWOC format.  A limited
number of other formats are available, depending on the print
program used.  In most cases the user may specify the length  of
the print line so that various paper sizes may be used.

A complete KWIC User Guide may be obtained by printing sequential
data set JUSD.KWIDOC (See Appendix H).


5.3.3    DSSUBMIT
                                                JAN 1988
A catalogued procedure is available to read a member of a
partitioned data set or a sequential data set and submit the
contents of that data set for processing as a standalone job

The following JCL is used to invoke the procedure =

     //STEPX EXEC DSSUBMIT,DSNAME='dsname1,FILE = wvvw,
     //           DSUNIT=uuuu,CCODE=xx
Where =
     dsname    The name of the data set to be processed.  If
               the data set is partitioned, the specific
               member must also be specified, such as

     vvvvvv    The volume serial number of the disk pack on
               which the data set resides.  This is not
               necessary for catalogued data sets.

     uuuu      The logical device type the data set resides
               on.  This is not necessary for catalogued
               data sets.

     xx        The condition code used to bypass the
               submission of the data set.  If any previous
               steps in the job return a condition code
               greater than or equal to the value of xx, the
               data set will not be submitted for
               processing.  If no CCODE is specified, the
               default is CCODE=M.
5. 3. "4
Initializing and Routing Plot Tapes
CalComp plotter output may be directed to either the Washington
Information Center or the NCC facility at RTF for plotting.  Each
center has a CalComp 1051 plotter.  Regional mailing of plotter
output is performed at both sites.

The general method for off-line plotting to either site is the
same.   A plot tape with appropriate header information and
optional plotting instructions must be initialized by the user
using  the NCC procedure PLINIT.  This plot initialization run
yields a specific tape volume serial number, assigned by the Tape
Management System, which must then be referenced in the CalComp
run(s) that follow.  The tape file must then be closed and the
volume catalogued.

NCC-IBM USER'S GUIDE                                     JAN  1988

Operations runs daily scheduled jobs that search the system
catalog for newly catalogued plot tapes. (Therefore, be sure  your
plot tapes are not catalogued before the plot information is
fully built.)  The MIC-destined plot tapes are transmitted via
telecommunications to WIC; the NCC-directed plot tapes are
plotted onsite.  Once the tape is transmitted or sent onsite  for
plotting, the catalog entry for that tape is deleted.  If the
user wishes to use the tape again, he must know the volume serial
number used (this can be found from previous job listings).

The high-order prefix of the data set name on the plot tape
determines at which site the plotting will actually occur.  All
plot tapes' names are of the form:

     RMT20.iiiaaaa.name          for WIC
     RMTO.iiiaaaa.name           for RTF

This name is formed through the PLINIT initialization program
using the symbolics BEST, ACCOUNT, and NAME plus the User-ID
found in the job name.

The user has the option of either cataloging his plot tape in a
separate step at the end of his job or he may catalog it within
the last step in which plot data is created.  To do this within
his last step he need only use CATLG in his DISP parameter; e.g.
DISP=(MOD,CATLG).   If he uses a separate step, the following  JCL
is sufficient:

      //step  EXEC PGM=IEFBR14
      //CATSTEP DD DSN=dsname-of-plot-tape,UNIT=TAPE,
      //   VOL=SER=vvvvvv,DISP=(OLD,CATLG)

where the data set name is of the form "dest.iiiaaaa.name" and
wvwv is the tape number assigned by PLINIT.

If the steps writing plot data to the tape are within the same
job as the tape initialization using PLINIT, the volume serial
number for the plot tape may be accessed through a backward
reference.   The data set name, however, must be given explicitly
and must agree with the name built from the PLINIT parameters:

   //ddname DD DSN=dest.iiiaaaa.name,
   //  VOL=REF=*.stepname.PLOTINIT.FT1MFOO1,
   //  DISP=(MOD,PASS)

where "stepname" is that name assigned by the user to the step
executing PLINIT.  If the user wishes to catalog the tape, the DD
card above should indicate DISP=(MOD,CATLG).

If the plot creation steps are in subsequent jobs, the volume
serial number of the tape is available from PLINIT's JES log


NCC-IBM USER'S GUIDE                                     OAN  1988

statements or the deallocation messages.

The JCL required for the use of PLIHIT is listed below.  Notice
however, that the DEST parameter will default to MIC if excluded.

  //stepname  EXEC PLINIT,
  //   ACCOUNT=ACCT,  (substitute your valid account code)
  //   NAME=PLOT,     (may be changed for unique qualifier)
  //   PRINT=A,       (default output from SYSPRINT)
  //   DEST=RMT20,    (indicate DEST=RMTO or =IOCAL for RTF)
  //   RETAIN=5       (default # of days that tape will be kept)
  //SYSIN  DD  *    (or //PLOTINST  DD  *)
      DELIVER TO   user name
                   user address

Note-' A total of eight delivery and special instruction cards may
      be input.

5.3.5    TAPESCAN

TAPESCAN is a utility which produces a summary dump of the
contents of a magnetic tape volume.  Only the first 132
characters of each tape block are printed.  TAPESCAN is
particularly useful when the data set names and characteristics
of the internal label of the tape are unknown.

Complete documentation on TAPESCAN is in the on-line data set
JUSD.TAPESCAN.DATA(DOCUMENT).   Several examples using the
standard TAPESCAN PROC for processing different types of tapes
are shown below.

1.  LIST summaries of the first 20 blocks of each file on an
   SL tape without a hex interpretation.


2.  SKIP 1 data file on an SL tape, LIST summaries of the
   first 75 blocks of the second data file.  (Each data file
   on a standard-label type has three actual

   to dump file x, SKIPTMn, where n = (3* (. x-1 ) ) +1 , MAXTMm, m = n+1

    //  EXEC TAPESCAN,VOL=vvvvvv,

3.  LIST summaries of the first 30 blocks of the first 2 files
   on an NL tape.


NCC-IBM USER'S GUIDE                                     JAN  1988

    //    PARM='LIST30,MAXTM2'

4. LIST the first 20 blocks of each file on a SPERRY tape.

    //    PARM='LIST20'

A full-screen TAPESCAN dialog utility is available through Option
E.U of the ISPF/PDF primary option menu.  An initial entry panel
is displayed for input of tape volser and pertinent tape
information.  When the user presses enter or return from the
first panel, a second panel is displayed for input of JOBCARD
information necessary to submit a batch TAPESCAN job.

5.3.6    Using Full-Screen Facilities to Advantage

Frequently users avoid using TCP or full-screen access methods
for the IBM 3090-300 because of the "screen refresh problem."
This problem occurs when the entire screen is "refreshed" or
completely rewritten for every ISPF transaction.  If the user
logged on at a relatively low baud rate (such as 2400 bpi or
less), the refresh is slow enough to be visible and, therefore,
frustrating.  At baud rates of 9600 bpi or more, the refresh
tends to be transparent.  This frustration can be avoided or
eliminated with practice.  In fact, at any baud rate, many of the
refresh operations can be entirely eliminated.  The result is a
significant increase in the speed of the work performed for the
user .    Stacking Commands

ISPF commands can be "stacked" by using the semicolon to separate
the commands.  For example, assume that you want to go from the
ISPF main menu to an edit session of the member $*LOGON in your
CLIST file.  Normally, you would enter the following commands
separately, and the screen would refresh after each time you
pressed the ENTER key.

     o  "2" to select the EDIT option from ISPF's main menu.

     o  Type in CLIST for other data set name on the Edit Entry panel

     o  Move beside the  member name $$LOGON and type "S" to select.

Since each screen refresh rewrites each line of your terminal
screen, this process is  slow and inefficient.  For a method of
accomplishing the same logical functions with only one screen
refresh, refer to the example in the discussion of MYED below.

NCC-IBM USER'S GUIDE                                     JAN  1988    Using MYLISTC

Note:  The following examples refer to CLISTS in the file
       JUSD.CLIST.  You can copy these into your ISO session
       by copying JUSD.CLIST($LOGON) into your file
       CLIST($$LOGON).   Watch the $ signs!  The next time
       you logon to TSO, the tools in JUSD.CLIST will be
       available to you.

JUSD.CLIST(MYLISTC) allows you to execute the TSO command LISTC
for a prefix of your choice.  It places the output of your LISTC
command into OUTPUTCTODAY),  a partitioned data set under your
system prefix.  MTLISTC then edits the catalog listing, stripping
away any irrelevant data trapped by the LISTC.  Control is
returned to you while editing OUTPUT(TODAY) with the ISPF Edit.

MYLISTC illustrates the following points=

     o  Output is sent to a disk data set.  This allows
        you to treat the output as data.  Results are not
        lost, and there is no waiting for batch processing.

     o  Use of ISPF Edit Macros.  The ISPF Edit macro
        JUSD.CLIST(EDLITC) performs a series of commands
        such as EXCLUDE ALL (X ALL), FIND ALL (F 'string1
        ALL) and DELETE ALL to remove the excessive data
        generated by the LISTC.    Using MYED

The CLIST JUSD.CLIST(MYED) allows fast entry to an ISPF edit of a
data set.  MYED is a stackable editor, i.e., edit sessions can be
"stacked" on top of each other without exiting from the current

The basis of the CLIST MYED is a statement as follows:

     ISPEXEC EDIT DATASET(data-set-name)

As an example, suppose that you want to examine the source code

Minimizing the number of screen refreshes, at the TSO READY
prompt, type the following:

     1.  ISPF 6

         This bypasses the ISPF menu panel.


NCC-IBM USER'S GUIDE                                     JAN  1988

         You are editing the PDS JUSD.CIIST and the member

     3.  GETSET

         Invoke the ISPF edit macro JUSD.CLIST(GETSET ).

Steps 2 and 3 can be combined in the following single command:


GETSET is used with those CLISTS which have certain lines that
can be found and labeled as per the ISPF Edit line label feature.
The system provides several default labels =

     o  .ZFIRST (or .ZF)

     o  .ZLAST (or .ZL)

     o  .ZCSR (for the cursor)

Labels are extremely useful for defining a vertical extent or
window of data which is to be processed by a block command (e.g.,
change, delete, or sort).

Now you can elect to go to the line which initiates the ISPF edit
of the output data set member created by MYLISTC:


This command finds the next line containing the string "ISPEXEC".

Another example illustrates the use of CLISTS and stackable
commands as follows :

     SPLIT;6,HYED 'JUSD.CLIST';s edlistc

This command splits your ISPF session, selects initial menu
option 6, invokes the CLIST MYED to edit JUSD.CLIST, and selects
member EDLISTC.  Only one screen refresh will occur.

Note:  Caution is required.  If the edit object is currently
       being edited by someone else, or by you in another panel,
       the dialog will error and some time may be lost.

Another use of MYED takes advantage of the stackable editor
functions.  You can go directly to edit from anywhere in an ISPF
session.  Suppose you are in a split ISPF session with both sides
busy.  You can type the following at the command line and not
lose either of your sessions:

NCC-IBM USER'S GUIDE                                     JAN 1988

     TSO MYED 'data-set-name';s member

To return to your split screen sessions, just enter the
following =


5.3.7    Using Group Data Sets

Many data sets have a 7-character prefix made up of the
concatenation of a User-ID and a prefix.  This is an effective
and common practice at NCC (and at other IBM sites), but it has
certain disadvantages both to the computer center and to the
user's management.  Group data sets, or data sets uith a
M-character prefix consisting of the account number, offer more
flexibility for the user and his management.

One major disadvantage is that when a user leaves employment, his
User-ID is removed from the list of valid users for that account.
Therefore,  all his data sets (prefixed by his User-ID) will not
be accessed and will eventually be removed from the system.  A
short-term solution to this problem is to copy or rename the data
sets of the terminating employee to another employee's User-ID.
This does nothing but prolong another tranfer when that employee

The recommended long-term solution is to convert long-term and
production type data set names from personal data sets (those
with 7-character prefixes) to group data set names based on the
account as the prefix.  By using group data sets, all the users
under the particular account can change, but the data sets need
not be renamed.
5.3.8    Using Partitioned Data Sets vs.  Sequential Data Sets

NCC recommends the use of partitioned data sets (PDS) for the
following reasons which benefit both the user and the NCC =

     o  Ease of use.

     o  Saves time.

     o  Saves money.

Partitioned data sets can hold many — even thousands — of
individual members.   Each one of these members can be viewed as
an individual sequential file.  Converting sequential files into
members of partitioned data sets offers the following benefits =

NCC-IBM USER'S GUIDE                                     JAN  1988

     o  Less likely to need to reload the sequential file.

     o  Less likely to lose the contents of the file.

     o  Use of the IBM-supplied utilities for processing
        PDS members (see OUSD.CLISTCALLMBRS) for an
        example of an ISPF Edit Macro).

NCC benefits also because it is more efficient to manage
partitioned data sets than to manage individual sequential files.

A PDS structure conserves more space than does a sequential file
structure.  For example, consider an 80-byte record in a
sequential file.  A track of disk space holds over 47,000 bytes,
and the minimum file size is one track.  Thus to create a
sequential file of one 80-byte record requires 47,000 bytes or
one track.  Most of this storage space is wasted.  The same data
can be stored in a partitioned data set, and the PDS can then be
used to hold more members.  As the number of members grows, the
structure becomes more efficient.

In another example, a user with 100 sequential files of one
80-byte record each and one track per file (or 100 tracks) has
4,747,6000 bytes reserved for the files.  The user must manage
these and ensure that they are accessed periodically in order to
be retained over time.  Note, however, that only 8,000 bytes are
actually being used to store data.  If these files are grouped
together with a common data set name and a member naming scheme
(such as iiiaaaa.project(file 1), iiiaaaa.rreject(file2),
iiiaaaa.project(fileS), etc.), the one file can be stored on one
track, an immediate savings of 99 tracks,  and can be easily

Output report files can also be converted to be members of
partitioned data sets.  Often report files are created as
sequential files with a record format that includes the A
attribute (e.g., FBA or VBA).  For many applications, these files
can be treated as members of a PDS which may then be reviewed
with the ISPF editor.  Using the ISPF editor to review a report
provides quicker decision making and quicker access to the report
contents than either SDSF or the 2 processor provides.   Saving
the report as a member of a PDS also allows the user to reprint
the report quickly and easily without rerunning the job that
created the report.

NCC-IBM USER'S GUIDE                                     JAN  1988

5.3.9    Using Date and Time Formats

The current date and time are available in many different
formats.  The current day of the week (e.g., MONDAY) is not
always directly available.  Many users have written CLISTS or
other programs to determine the day of the week.

The data set JUSD.DAILY.STATS(TODAY) contains the date and time
in many different formats including Julian, American, and
European date formats, and the day of the week.  This member  is
updated shortly after midnight (usually within  10 seconds) of
every production day at the NCC.  On Sunday nights the IBM 3090
is offline at midnight, so the update occurs at IPL (usually  by
0630).  Note that the member TODAY is created by the FORTRAN
program documented in the first data line of the member.

TODAY allows for an automated method of retrieving the data and
time including the day of the week, a useful way for routine  or
recurrring production jobs which must determine the day of the
week.  For example, production jobs may run Monday through Friday
with no processing on Saturday or Sunday.  The user must find a
reliable way to determine the day of the week in order to prevent
processing on Saturday or Sunday.

Reading the data set JUSD.DAILY.STATS(TODAY) allows the user  to
get the current day of the week in a standard manner.  The member
YESTERDA contains the previous day's information.

S.3.10   Interactive Processing vs. Batch Processing

While many users may continue to use batch processing facilities,
NCC expects that many other users will develop and execute their
applications for interactive processing under TSO.  In support of
this expectation, NCC does not currently charge a premium for
using TSO.  Or in other words, the preferred (PRF) service cost
algorithm is set to 1.0 for TSO.

                                         T»    Advantages of Interactive Processing

NCC anticipates a move from batch processing to interactive
processing and has directed its support efforts in this
direction.  Many advantages to TSO processing have been

NCC-IBM USER'S GUIDE                                     JAN  1988

     o  Expanded telecommunications facilities.  The NCC's expanded
        telecommunications network is designed to support high-speed,
        interactive processing in the TSO environment.  In particular
        network design allows as many users as possible to use the
        Interactive System Productivity Facility (ISPF).  For
        example, ISPF's Editor allows data to be sorted immediately.

     o  Execute programs directly.  TSO allows the user to execute
        programs interactively and thereby avoid delays in batch
        queues.  For example, SAS and FOCUS both support inter-
        active processing.  The system CLIST for SAS--
        SYS2.EPA.CLIST(SAS)—allows the TSO user to specify the SAS
        commands,  input data, and output reports through optional

     o  fiuick processing of small jobs.  Small, quick jobs (usually
        3 CPU seconds or less) are candidates for interactive
        processing.   (Note:  NCC does not support the TSO TAPE
        MOUNT capability so jobs that use a tape must be processed
        in batch mode.)

     o  Execute TSO in batch environment.  TSO commands in the
        foreground (interactive) are almost exactly the same as
        TSO in batch.  Therefore, the user has little difficulty
        converting from interactive to batch.  Conversely, batch
        users must learn the CLIST language in order to run their
        programs in TSO interactively.

     o  NCC sponsored classes.  NCC encourages users to become
        proficient in TSO and CLIST construction by sponsoring

     o  IBM special features.  IBM software often allows the
        full-screen TSO user to become more productive by using
        special built-in features.  For example, the IBM ISPF Edit
        allows an immediate SORT; the IBM line editor (IED) does
        not.    Example of a CLIST

The following example of a CLIST depicts a program written in
FORTRAN and executed interactively.  The program has already been
compiled, linked, debugged, and saved in the departmental load
library (DPT1.LOAD) as member PROG1.  It is automatically made
available to any user in the department during the TSO logon
process because it is a CLIST set up as a group data set.  The
program uses two input files and creates one output file.  One
input file is the department's data base and the other is
supplied by the user.  The output file can be directed to the

NCC-IBM USER'S GUIDE                                      JAN 1988

The CLIST DPT 1 .CLIST(PRG1) is as follows:

     FREE FILE(FT11F001,FT12F001,FT06F001,FT05F001)

     WRITE Enter 1 to direct the output  of  PRG1  to  your terminal.
     WRITE (All other values direct  the  output  to  a DASD dataset)

     SET Sout = £STR(£SYSPREF. .LIST 133(PRG1))
     READ Canswer
     IF £ans Efi 1 THEN SET £out = £STR(*)

     AILOC FKFT1 1F001 )  DA ('DPT 1 . PRG 1 . DATABASE')  SHR
     ALLOC FKFT12F001)  DA ( ' UUUDPT 1 . PRG 1 . USERDATA ' )  SHR
     ALLOC FKFT05F001)  DA(*)
     ALLOC FKFT06F001)  DA(£out)


     EXIT (filastcc)

NCC-IBM USER'S GUIDE                                     JAN 1988


The NCC maintains many different programming language compilers,
interpreters, and assemblers for EPA users.  Selection of one
language over another for a particular job is usually dictated by
the type of task being programmed.  For instance, a scientific or
mathematical task would normally be coded in the FORTRAN or PL/1
language, whereas COBOL or PL/1 would normally be chosen for
commercial systems.   The choice of a programming language should
also be in consonance with the skill of the programmer involved
since learning a new programming language by trial and error can
be expensive and time-consuming.

Compilers are software systems which accept statements coded in
specific languages as input and produce object code as output.
During the compilation process, syntactical errors are noted
automatically and many other useful programmer aids can be
produced on option.   Compilers are used exclusively to produce
machine-executable code from high-level language source
statements.  High-level languages are so named since each
language statement will produce many machine language
instructions when compiled.

Interpreters also operate on high-level source languages.  They
are significantly different from compilers since they interpret
and execute each statement as it is encountered.  Consequently,
interpretive languages are generally most economical for one-shot
programs or time-critical development work.  Programs which are
to be run repeatedly would be far too expensive to run
continually in this  mode.

Assemblers process low-level languages with instruction sets
which have almost a one-to-one correspondence with the machine
instructions they produce.  Such languages are usually used for
systems software coding or for coding which has to be as
efficient as possible.

The primary output produced by compilers and assemblers is object
code.  Object code cannot be executed directly.  To produce
executable code, object modules must be processed by a systems
software package called the Linkage Editor, which is an IBM-
supplied utility package.   The Linkage Editor can accept one or
more object modules  as input and produce executable or
intermediate load modules as output.  Intermediate load modules
are not executable and are usually produced for programs which
are part of a larger system that will later be input to a Linkage
Editor run to produce an executable load module from several
intermediate load modules.  Intermediate load modules are smaller
since they do not include any language subroutines or external
routines.  It is for this reason they cannot be executed.  The
Linkage Editor can also use intermediate or nonexecutable load

NCC-IBM USER'S GUIDE                                     JAN  1988

modules as input and will accept a mix of object and load modules
as input.  Generally, the user will not be concerned with the
complexities of the Linkage Editor since the procedures, which
have been supplied for language users, automatically invoke this
system.  However, this package is explained fully in the IBM
publication entitled OS/VS Linkage Editor and Loader (GC26-3813).
An abbreviated explanation on the use of the Linkage Editor is
covered in Section 5.3.9.  The programming languages which the
NCC supports are as follows =

     o   Assembler Language,  the lowest level language for
         use on the IBM 360/370 series of computers.

     o   COBOL, a high-level language which is generally
         used for programming of business and commercial
         applications which do not involve an inordinate
         amount of computation.

     o   FORTRAN and APL, high-level languages which are
         used mainly for scientific and mathematical

     o   PL/1, a high-level language which contains, in a
         single language, many of the attributes of
         Assembler Language,  COBOL and FORTRAN.

     o   BASIC, an interpretive high-level language with
         limited computational and manipulative functions
         which is easy to learn and use.

See Appendix G for a summary of all software available and
supported at the NCC-IBM.

The NCC has established standards governing the construction  and
use of language processor catalogued procedures.  Step names  and
commonly used symbolics are given consistent names shown below.
Any procedure which invokes a compiler, assembler, or interpreter
must be supplied with the following JCL card:

         //xxxx.SYSIN DD

This DD statement must point to the data set containing the
source program or is followed by the source program in the input
stream.  xxxx is the name of the compile step; e.g., FORT for
FORTRAN, COB for COBOL etc.  Similarly, any input into Linkage
Editor or GO steps should be included in order as //LKED.SYSLIN
DD and //GO.whatever DD respectively.  All printed output for
each procedure is controlled by its symbolic parameter PRINT.
This parameter has a default value of A.   All punched output  is
controlled by the two symbolic parameters DECK and PUNCH.   DECK
has the default value DUMMY and PUNCH has the default value B. To

NCC-IBM USER'S GUIDE                                     JAN  1988

obtain punched output, the user should specify (on the EXEC
statement) DECK=SYSOUT and PUNCH=x, where x is the appropriate
SYSOUT class.

The sections which follow contain brief descriptions of each  of
the language processors listed above, provide details of the
procedures which have been provided for their use, and describe
the publications which are available for users requiring
additional information.

Assembler Language (ALC, from Assembler Language Coding) is a
low-level symbolic programming language for use on the IBM
360/370/30xx series of computers.  ALC instructions have almost a
one-to-one correspondence with absolute machine instructions.
However,  ALC does allow macroinstructions to be created and
stored for inclusion into any number of ALC programs.  This
facility tends to make ALC programming more reasonable over a
period of time.  ALC is used for most systems software work since
it allows the programmer to code at the same level as the
computer operates.  Because of ALC's micro nature, it is rarely
used for the development of application systems since the cost of
such developments and of the subsequent maintenance can be cut
substantially by the use of a high-level language.

6.1.1    Catalogued Procedures

The NCC provides and supports several catalogued procedures for
ALC.  All ALC procedure names are of the form:


where v is the version (H) and fff is an acronym for the function
performed by the procedure.  The procedures are=


     Assembler HV2




                                                         JAN 1988

                     assemble only

                     assemble and

                     assemble, link,
                     assemble and execute
                     without link-edit

                     alias for ASMGCG

           —        link-edit and

Each of the procedures listed above includes symbolic parameters.
These have been assigned defaults which may be overridden by the
user.  The parameters included,  their defaults, and meanings are =


Parameter     Default

PRINT         A










                         JAN  1988


SYSOUT class for printed

UNIT for object deck (DUMMY
causes none to be punched;
if an object deck is
desired, the value of
'SYSOUT=Bf must be
provided. )

FARM field for the assembler

The region available to the

The region available to the
Linkage Editor

FARM field for the Linkage Editor
The region available to the
compiled program

The maximum number of bytes
available to the Linkage
Editor (used as the first
component of the SIZE

The number of bytes to be
reserved for buffers by the
Linkage Editor (used as the
second component of the SIZE

Means that the loader may
use as much space as is
available in the region size
specified for its own code,
access methods, buffers and
tables, and the loaded

NCC-IBM USER'S GUIDE                                     JAN  1988

6.1.2    Publications

Further information concerning parameters which may be supplied
for ALC assemblies and Linkage Editor runs can be found in the
publications listed below.  These publications are all concerned
with either ALC or the Linkage Editor and Loader.

     o   IBM System/370 Extended Architecture Principles of
         Operation (SA22-7085)

     o   Assembler H Version 2 Application Programming Guide

     o   Assember H Version 2 Application Programming
         Language Reference (5026-4036)

     o   MVS/XA Supervisor Services and Macro Instructions

     o   MVS/XA Data Administration Guide (GC26-4013)

     o   MVS/XA Data Administration:  Macro Instruction
         Reference (GC26-4014)

     o   MVS/XA VSAM Administration Guide (GC26-4015)

     o   MVS/XA VSAM Administration:  Macro Instruction
         Reference (GC26-4016)

     o   MVS/XA Linkage Editor and Loader User's Guide
         (GC26-401 1 )
6.2    COBOL

COBOL (COmmon Business Oriented Language) is a high-level
programming language, with English-like syntax, which is used
mainly for commercially-oriented application programming.  Since
COBOL is used widely, the American National Standards Institute
published standards which were intended to create a language
which could be used interchangeably on and by all computer
systems which offer COBOL.

When compiled, most COBOL instructions cause a large number of
machine language instructions to be generated.  However, the
COBOL programmer need not be a skilled Assembler Language
programmer in order to use this language.

NCC-IBM USER'S GUIDE                                     JAN  1988

The versions of COBOL supported by the NCC are the IBM-supplied
OS/VS COBOL and VS COBOL II.  VS COBOL II is the latest  software
technology.  OS/VS COBOL is expected to be removed by June  30,
1988.  The NCC also supports Optimizer III, a COBOL optimizing
package (see Section 7.6.1) and COBOL conversion aid, a  set of
programs and commands to assist with conversion of OS/VS COBOL
source code into VS COBOL II.  Information on the COBOL
conversion aid can be found in data set JUSD.VSCOBOL2.DATA.
6.2.1    Catalogued Procedures

The NCC provides and supports several catalogued procedures  for
COBOL users.  All COBOL procedure names are of the format:

     COBUfff - for OS/VS COBOL
     COB2ufff - for VS/COBOL II

where fff is an acronym for the function performed by the
procedure.   The procedures are =

    VS COBOL II   OS/VS COBOL*        Function

     COB2UC        COBUC         compile only
     COB2UCL       COBUCL        compile and link-edit
     COB2UCLG      COBUCLG       compile, link-edit, and execute
     COB2UCG       COBUCG        compile, load, and execute
     COB2ULG       COBULG        load and execute
     COB2UCX       COBUCX        alias for COB2UCG and COBUCG

     ^Planned removal in summer 1988.

Each of the above procedures includes symbolic parameters.   These
parameters  have been assigned default values which may be
overridden by the user.  The parameters, their defaults, and
meanings are =





















                                                         OAN  1988


                                SYSOUT class for printed output

                                SYSOUT class for object deck

                                FARM field for the COBOL compiler
                                which has a default of

                                The region available to the

                                The region available to the Linkage

                                The region available for execution
                                of the compiled program

                                PARM field for the Linkage Editor
                                FARM field for the Loader

                                PARM field used in the GO step
                                of the COBUGLG procedure.

                                The maximum number of bytes
                                available to the Linkage Editor

                                The number of bytes to be reserved
                                for buffers by the Linkage Editor

                                Specifies that the Loader can use
                                all remaining space in the region
                                for itself, the loaded program,
                                access methods, and any buffers and
                                tables needed.

Optimizer Object Code can be requested by specifying OPTIMIZE as
a parameter.   This results in considerably smaller object
programs than are produced without optimization.  For COBOL
programs that are not I/O bound, execution time is reduced.
Further information concerning parameters which may be supplied
to the COBOL compiler or to the Linkage Editor is contained in
the publications listed below.

NCC-IBM USER'S GUIDE                                     JAN  1988

     o   IBM VS COBOL for OS/VS2 (GC26-3857)

     o   IBM OS/VS COBOL Compiler and Library Programmers
         Guide (SC28-6483)

     o   OS/VS Linkage Editor and Loader (GC26-3813)

     o   VS COBOL II Application Programming Guide
         (SC26-4045-1 , Release 2)

     o   VS COBOL II Application Programming;  Debugging
         (SC26-4049-1 , Release 2)

     o   VS COBOL II Application Programming:  Language
         Reference (GC26-4041-3, Release 2)

     o   COBOL Conversion Aid (SH-20-6402)

6.2.3    VJ5 COBOL Migration

Over the next several years, users can expect to see the demise
of many software products that do not support 31-bit addressing.
Many of these products already have their replacement version
available today.  COBOL is one such product.  The VS COBOL II
compiler supports 31-bit addressing.  It has been available for 3
years and is already in its second release.  If you have not
started migration to VS COBOL II, now is the time to begin.  If
you have the new COBOL installed but have not fully implemented
it, (i.e.,  deleted the old VS COBOL compile and library), it is
time to set a target date for completing the migration to VS

The 31-bit addressing support is not the only advantage of VS
COBOL II.  Many performance and functional enhancements have been
incorporated.  For example, programs compiled under VS COBOL II
are reentrant.  The old VS COBOL probably will not receive any
development attention in this area.  In fact, VS COBOL may be
"functionally stabilized" in the near future.

The NCC recommends that the migration alternatives be reviewed
and that at least the following actions be taken:

     o  That all new COBOL development be done with VS COBOL

     o  That any application procured or migrated from
        outside sources be converted to COBOL II.

NCC-IBM USER'S GUIDE                                     JAN  1988

     o  That when significant changes are required to
        existing COBOL programs, the programs first be
        converted to COBOL II and then the changes applied.

NCC promotes the use of the most efficient compilers and attempts
to prevent unsupported software from being used by production
applications.  Users will be kept informed about the future
status of VS COBOL.

COBOL Conversion Aid allows users to convert OS/VS COBOL and CICS
command source code to VS COBOL II source code.  The COBOL
conversion aid can be run interactively or in batch.

Before using the conversion aid, each user must establish a one-
time special environment.  For the JCL needed to create the
special environment, see JUSD.COBAID.DATA(CONVERSN).

A list of available IBM manuals which describe the COBOL
conversion aid is in JUSD.COBAID.DATA(MANUALS).

The key facilities provided by the COBOL conversion aid are as
follows =
     o  Ability to convert most syntax differences between
        OS/VS COBOL and VS COBOL II programs.

     o  Removal and/or conversion of the Base Locator for
        Linkage (BLL) section mechanism and reference.

     o  Conversion of both source programs and copy modules.

     o  Conversion of management reports.

     o  A statement-by-statement diagnostic listing showing
        by program the result of the conversion process.

     o  A facility for changing and/or creating COBOL
        conversion modules.

     o  Elimination of conflicts between v.ser-defined names
        and words reserved for VS COBOL II.

     o  Open converter design able to perform conversion
        from various levels of COBOL into other COBOL

NCC-IBM USER'S GUIDE                                     JAN  1988

Major enhancements added to the COBOL conversion aid are as

     o  Support for Floating Point usage.
     o  Support for Exponentiation usage.
     o  Support for Complex Occurs Clause usage.
     o  Support for Sort Special Registers usage.
     o  Support for Tally Special Register usage.
     o  Support for Apply Write-Only usage.
     o  Conversion of EXEC CICS commands.
6.3    FORTRAN

FORTRAN (FORmula TRANslator) is a high-level programming language
which is used primarily for scientific, engineering, and other
applications which involve extensive mathematical computations.
This language is available on most computer systems throughout
the world and, consequently, the American National Standards
Institute has defined a standard FORTRAN language.

Several preceded program and subroutine libraries are available
to NCC FORTRAN users.  Using these libraries, complex programs
can be constructed quickly and inexpensively.

VS FORTRAN is the only FORTRAN compiler supported at NCC and has
the following components =

     o  Compiler.  Translates FORTRAN source code to object code
        for subsequent execution with the support of the FORTRAN

     o  Library.   Contains mathematical, character, bit, service,
        input/output, and error routines in support of the compiler

     o  Interactive Debug.  Allows the programmer to monitor the
        execution of FORTRAN programs and to examine the change
        data at execution time.

The compiler is available interactively via ISPF or in batch via
catalogued procedures.  Additionally, Interactive Debug can be
executed in line mode.

JUSD.VSFORT2.DATA is the on-line data set providing more
information on this product.

6.3.1     Catalogued Procedures

The NCC provides and supports several catalogued procedures for
VS FORTRAN.  The VS FORTRAN procedures are'
                              6-1 1


     VS FORTRAN   Functions
                                                JAN  1988
     FORTVC      compile only
     FORTVCL     compile and link-edit
     FORTVCLG    compile, link-edit, and execute
     FORTVCG     compile, load, and execute
     FORTVLG     link-edit and execute
     FORTVL      load and execute
     FORTVG      execute only
     FORTRC      re-entrant compile only
     FORTRCL     re-entrant compile and link-edit
     FORTRCLG    re-entrant compile, link-edit, and execute
     FORTRLG     re-entrant link-edit and execute
     FORTVIAD    execute InterActive Debug in batch.

Each of the procedures listed above contains symbolic parameters
These have been assigned defaults which may be overridden by the
user.  The parameters, their defaults, and meanings are
documented in the on-line data set JUSD.VSFORT2.DATA(PROCS).
Further information concerning parameters which may be supplied
to the FORTRAN compiler or the Linkage Editor and Loader can be
found in the publications listed below.  These publications are
all concerned with either the FORTRAN language or the Linkage
         VS FORTRAN Version 2•
                       General Information
         VS FORTRAN Version 2 =   Language and Library
         Reference (SC26-4221)

         VS FORTRAN Version 2-   Programming Guide

         VS FORTRAN Version 2-   Interactive Debug Guide and
         Reference (SC26-M223)

         VS FORTRAN Version 2:   Reference Summary

         OS/VS Linkage Editor and Loader (GC26-3813)

         MVS/XA Linkage Editor  and Loader (GC26-4011)

NCC-IBM USER'S GUIDE                                     JAN  1988

6.4    PL/1

PL/1 (Programming Language 1) is a multipurpose, high-level
language which lends itself to commercial and scientific
applications as well as to real-time and systems software work.
The language contains many of the capabilities of FORTRAN, COBOL,
and ALGOL as well as some of the attributes of Assembler
Language.  The NCC supports the PL/1 Optimizing Compiler.

6.4.1    Catalogued Procedures

The NCC provides and supports catalogued procedures for the PL/1
optimizing compiler.  The PL/1 procedures are=

          PL/1 Optimizing       Function

           PLIXC                 compile only
           PLIXCL                compile and linK-edit
           PLIXCLG               compile, linX-edit, and
           PLIXCG                compile, load, and execute
           PLIXLG                load and execute; translate
                                 and interpret

The procedures listed above contain symbolic parameters.  Most of
these parameters have been assigned defaults which users may  wish
to override.  The parameters included, their defaults, and
meanings are as follows =


Parameter     Default

PRINT         A








                         JAN  1988


SYSOUT class for printed output

For object deck (DUMMY causes none
to be punched; if an object deck is
desired, the value of  'SYSOUT=B' must
be provided . )

PARM field for the PL/1 compiler
which has a default of

The region available to the

The region available to the Linkage

Defines the transient routine
library to be used.   Should be set
to TASK if multitasking is being
used .

PARM field for the Linkage Editor
PARM field for the Loader

The region available to the
compiled program

The amount of core available to the
Linkage Editor

The amount of core available for
Linkage Editor buffers

Parameters for the execution of the
compiled program

Means that the loader may use as
much space as is available in the
region size specified for its own
code, access methods, buffers and
tables, and the loaded program.

NCC-IBM USER'S GUIDE                                     JAN 1988

6.4.2    Publications

Further information concerning parameters which may be supplied
for PL/1 compilations and Linkage Editor and Loader runs can be
found in the publications listed below.  These publications are
all concerned with either the PL/1 language or the Linkage Editor
and Loader.

     o   OS PL/1  Optimizing Compiler:  Programmers Guide

     o   OS PL/1  Optimizing Compiler:  Messages (SC33-0027)

     o   OS/PL/1  Optimizing Compiler:  TSO User's Guide

     o   OS PL/1  Language Reference Manual (GC26-3977)

     o   IBM OS Linkage Editor and Loader (GC26-3813)

     o   MVS/XA Linkage Editor and Loader (GC26-4011)

6.5    BASIC

BASIC (Beginners' All-purpose Symbolic Instruction Code) is an
interpretive language which was originally developed at Dartmouth
College as a training aid for student programmers.  The version
available at the  NCC, IBM's VS BASIC, is fully supported.

BASIC is an inefficient language for production runs.  The reason
is that, being an interpreter, each statement is compiled
immediately prior to its execution.  Its primary benefit is the
speed and ease with which it can be learned and applied to a
specific problem  by a nonprogrammer.

Since BASIC is an interpreter, there are no catalogued procedures
available to NCC  users.

The following VS  BASIC publications are available at the NCC
Technical Library.

     o   VS BASIC Language (GC28-8303)

     o   VS BASIC OS/VS and DOS/VS Programmers Guide

NCC-IBM USER'S GUIDE                                     JAN 1988


An application package is generally a software system that
provides either a powerful utility function such as a Data Base
Management System (DBMS) or one that performs a specific task,
such as statistical analysis.  A utility-type system such as a
DBMS enables data to be stored without redundancy while
programmers are relieved of the repetitious coding of I/O
routines.  A statistical analysis package, on the other hand,
would most likely require only control statements and data to
produce results that would otherwise take many work-months of
programming time to duplicate.

The application packages that are available to NCC users are
those that best meet the needs of the EPA user community.  These
packages were developed by hardware vendors, software vendors,
computer service utilities, colleges and universities, and
government agencies.   Some of the packages are available for only
a nominal fee; others are expensive and have complicated
contractual arrangements.

To ease the use of the application packages, certain standards
have been established.  These standards cover the naming of the
load module libraries containing the programs, the use of common
conventions in catalogued procedures provided for the packages,
and the user procedure library.

An application package may consist of one or more executable
programs.  Each package has its own library reserved for programs
which are a part of that particular package.  The library name
consists of word(s) identifying the package, the version (if
applicable), and sometimes other identifying information.  The
libraries containing application package programs all reside on
disk packs which are permanently mounted (on-line).  Therefore,
the facilities provided by these packages are available whenever
the computer system is available.

Catalogued procedures have been developed for most of the
application packages discussed in this section.  These procedures
are maintained in a System Procedure Library by NCC personnel.
Even though each procedure serves a different purpose, they all
share the following characteristics:

     o   The name is a derivative of the name of the

     o   The appropriate STEPLIB DD statements are included.

NCC-IBM USER'S GUIDE                                     JAN  1988


7.1.1    ADABAS Data Base Management System

ADABAS is a Data Base Management System (DBMS) developed, and
supported by Software AG.  ADABAS provides system developers with
an efficient, flexible, and comprehensive capability for meeting
data management requirements for a variety of application
environments.  EPA has installed the ADABAS data base management
system to meet the Agency's "central" DBMS (i.e., the DBMS
software is shared among multiple users and applications)
requirements and has designated ADABAS as the Agency's standard

ADABAS can access data using multiple "keyed" selection criteria,
supports complex data structures, conserves data base storage
space through the use of data compression techniques, provides
excellent backup, recovery, and data security facilities, and
provides a comprehensive set of utilities to easily accommodate
data base structural changes with minimal impact on existing
programs.  ADABAS can be used in both batch and interactive
processing modes and can interface with application programs
written in COBOL, FORTRAN, PL/I, and Assembler languages.

The ADABAS DBMS provides application developers and users with
two tools for interfacing to application data bases.   NATURAL is
a nonprocedural language that can provide end users and
programmers with the capability of querying, reporting on, and
updating ADABAS data bases.  It can operate in both batch and on-
line modes.   The ADABAS direct call facility provides the
application programmer with a procedural interface to ADABAS;
however, its use must be explicitly approved by the Central Data
Base Administrator (CDBA).

The following references are suggested for the user=

     a.  ADABAS Application Development Procedures Manual

     b.  Guidelines for Data Base Design in the ADABAS

Note:  These manuals may be obtained by calling the Data Base
Support Section at (919) 5M1-M262 or (FTS) 629-4262.

     c.  A list of ADABAS manuals, a brief summary of their
         contents and ordering instructions may be obtained
         by accessing data set JUSD.ADABAS.DATA(DOC).

NCC-IBM USER'S GUIDE                                     JAN  1988

7.1.2    SYSTEM 2000

SYSTEM 2000 is a general-purpose data base management system
which allows a user to define, load, retrieve, and update data
bases.  The system uses its own elementary command language and
supports fully nested Boolean logic for selection criteria.
Functions and command strings can be defined by the user and
stored for later use.  Addressing and I/O operations are handled
automatically for the user.  The system also includes a report
writer, optional audit trail facility for updates, and an
interface which permits COBOL, FORTRAN, or PL/1 programs to be
used with the data base.

Each SYSTEM 2000 data base is composed of six BDAM files.  If an
audit trail is desired, an optional update file is available.
With SYSTEM 2000 batch, remote, or conversational use of the
files is possible.  The command language syntax provides a
complete set of commands for defining, restructuring, saving,
sharing, and releasing a data base, as well as for updating and
retrieving data.  Security provisions may be used to restrict
access to the data base for retrieval and/or update purposes.
Data security may be maintained by password control to the data
base and by additional passwords for each component.  SYSTEM  2000
also has several built-in statistical functions such as sum,
count, mean, maximum, minimum, and standard deviation.

Three features of SYSTEM 2000 are described below;

     a.  The Programming Language Extension (PLEX) enables
         manipulation of data in a SYSTEM 2000 data base
         from a programming language such as COBOL, FORTRAN,
         or PL/1.  This feature provides the mechanism for
         addressing any part of the data base of interest to
         the "host" program, to retrieve data, and to update
         the data base from the program.

     b.  The ACCESS feature provides a user-oriented
         language with which a nonprogrammer may express his
         request for retrieval or updating of a data base.
         The system is easy to learn and is highly suited
         for interactive use from remote keyboard terminals.

     c.  The Report Writer feature enables the user to
         prepare report definitions such as page heading
         specifications, page ejection, and the accumulation
         of sub or grand totals dynamically via conditional
         statements.  Up to 100 reports can be generated
         from a single pass of the data base index files.

Other features of SYSTEM 2000 are discussed in the manuals listed

NCC-IBM USER'S GUIDE                                     JAN  1988    Publications

SYSTEM 2000 Language Specification Manuals (LSM's) and SYSTEM
2000 Messages and Codes are suggested as references for the user.
For a complete list of current publications on SYSTEM 2000 which
are pertinent to this site, as well as more inclusive information
on SYSTEM 2000 usage and JCL requirements, refer to the on-line
data set JUSD.S2K.DATA (see member CONTENTS).    DBACCESS

The DBACCESS software package gives COBOL PLEX programmers in the
SYSTEM 2000 environment the capability to dynamically issue the
LOCATE command with full WHERE CLAUSE processing.  This means
that the COBOL PLEX program, when coded accordingly, can be
executed as a typical load module and modify or create the
appropriate LOCATE commands as required for each user application
of that COBOL program.

For complete details on coding for DBACCESS and its usage
requirements, the user may refer to member DBACCESS in the on-
line documentation in the data set JUSD.S2K.DATA.

7.1.4    Battelle's Automated Search Information System (BASIS)

BASIS is a Textual Information Management System (TIMS) from
Information Dimensions, Inc., a subsidiary of Battelle.  The data
base environment is such that there is only one data base for the
data base environment.  There are many utilities with which users
and data base analysts may query and manipulate the data.  BASIS
was designed to be a superior tool for managing and using large
text data bases.   It has been particularly useful to the EPA in
the areas of bibliographic/library systems and litigation
support.  Other clear choices for BASIS applications include
technical information lookup, newspaper and journal archiving,
law and regulation tracking, and records management.

BASIS has some exceptionally useful retrieval capabilities.
Among them are the ability to recognize plurals of searched-for
words.  For example, if you were searching for "alumnus," BASIS
(if so directed)  would also find "alumni."  BASIS can also do
synonym searching through its Thesaurus feature.  For example, if
you were searching a set of documents for all references to
"Belize," BASIS would also search for "British Guiana."  Another
aspect of the Thesaurus feature is the ability to standardize
data within the data base.  For example,  one might allow state
abbreviations to be input, but full state names would be stored
in the data base.

NCC-IBH USER'S GUIDE                                     JAN  1988

Documentation on BASIS can be found in the on-line data set

Several points of contact have been established for BASIS
support =

     o  For questions and problems with the day-to-day
        operation of the Data Center as they relate to an
        existing BASIS application, call NCC User Support
        (see  Appendix A for telephone number).

     o  For information about the content and availability of
        existing BASIS applications, con-L^ct the Data Base
        Administrator for that application.


7.2.1    Inquiry and Reporting System (IRS)

The Inquiry and Reporting System is an information retrieval
system designed to assist management in acquiring needed
information from computer data banks.   This package offers many
of the capabilities associated with the higher-level languages
such as COBOL, without the intricacies involved in the use of
these languages.  IRS serves users by allowing them to access
information and prepare one-time reports with a minimum of delay.
The programmer can also use IRS to prepare recurring reports.

One of the prime advantages of IRS is its ease of use.  Users
with a minimum knowledge of data processing can develop and
implement simple reports.  IRS is written in Assembler language
and employs interpretive and generative techniques.   IRS
interprets the parameters generating logic table code for
immediate execution of the request.  Intermediate compiles and
processing are not required.

The system should not be used to update data files or for
mathematical  computations and analysis beyond tabular reports.
Users encounter limitations that seem to be system problems in
the following areas =

     o   Use  of the tabulation feature with sum line

     o   Use  of automatic date and page placement on a
         footing line

Some of the most important features of IRS include the following:

     o   READ multiple input files, physical or indexed

NCC-IBM USER'S GUIDE                                     JAN  1988

         sequential, from several devices simultaneously.
         Records can be fixed or variable length.

     o   RETRIEVE records meeting user specifications.
         Record selection based upon simple compare
         statements using AND, OR, and ELSE connectors.

     o   ACCESS character, packed decimal (IBM only) or
         binary data by either Start Position or Field Name.

     o   PRODUCE multiple outputs in one pass of the input

     o   SORT records in ascending and/or descending

     o   CHAIN to data from auxiliary files in the
         selection,  sorting,  and output operations.

     o  - GENERATE printed reports and summaries tailored to
         user specifications.

     o   FORMAT reports complete with report title, date,
         columnar headings, footings, and page number; uses
         full complement of spacing and editing options such
         as floating dollar sign, zero suppression, page
         overflow, etc.

     o   SUMMARIZE data including the specification of
         several levels of totals and a full complement of
         spacing and editing  options such as counts,
         averages, maximum values, etc.

     o   PRINT reports on a variety of standard and
         nonstandard forms such as labels, checks, invoices,
         purchase orders, ledger cards, etc.

     o   CREATE individual, combined, or collated files from
         selected data on magnetic tape, punched cards, and
         direct access devices.

     o   RECURRENT comparisons and indexed moves may be
         performed within the Select/Sort Module.

     o   EXIT to a FORTRAN or COBOL subroutine for more
         efficient computations or interface with data base
         management systems.

The following references are  suggested for the user'-

     a.   Inquiry and Reporting System User's Guide

NCC-IBM USER'S GUIDE                                     JAN  1988

     b.  IRS Training Supplement:  Sample Problems and

     c.  Data set JUSD.IRS.DATA has examples typical of IRS

EASYTRIEVE PLUS is an information retrieval and data management
system designed to simplify computer programming.  Its English-
like language and simple decalarative statements provide the new
user with the tools needed to produce comprehensive reports with
ease, while its enhanced facilities provide the experienced data
processor with the capabilities to perform complex programming

EASYTRIEVE PLUS can be used f or =

     o   Report generation

     o   Information retrieval

     o   Record updating/editing

     o   Matching/merging files

     o   Preprinted or customized forms or microfiche output

The EASYTRIEVE PLUS package can access sequential, ISAM, VSAM, or
data base data sets on disk or tape.  It can handle fixed,
variable, undefined, or spanned records.  Editing, sorting,
logical file searches, and report generation are other features
of the package.  The EASYTRIEVE PLUS system can interface with
most data base systems, extract information from any storage
facility, and work with data in alphabetic, numeric, packed-
unsigned, or binary form.  EASYTRIEVE PLUS accepts any number of
input files in a single job when processing keyed files .

EASYTRIEVE PLUS reports may be generated with or without a sort,
depending on the application.  Sorting of files is also available

Users may access EASYTRIEVE PLUS by using the catalogued
procedure EASYTPLS.  This procedure supplies the appropriate GCL
for all EASYTRIEVE PLUS functions.

Documentation is available on-line in JUSD.EASYTREV.PLUS.DATA.

NCC-IBM USER'S GUIDE                                     JAN 1988


7.3.1    Biomedical Computer Programs

The Biomedical Computer Programs are a comprehensive and
integrated library of statistical programs.  These programs may
be loosely classified into six series:  Data Description,
Frequency Tables, Regression Analysis, Analysis of Variance,
Multivariate Analysis, and Special (including nonparametric
statistics and cluster analysis).

BMDP offers the user a broad set of descriptive and analytical
procedures.  These procedures have convenient defaults, offer
supplementary printer-graphics, and have thorough, comprehensive
documentation.  Data grouping or selection may be affected
through the control language.  The generation of new variables,
as well as data modification, can be accomplished through the use
of a control language or the inclusion of a FORTRAN subroutine in
the BMDP run.

Those users with sufficient statistical knowledge to select any
appropriate BMDP program will have little difficulty implementing
the control options available in each program, whether sizing
histograms or choosing a method of initial factor extraction.

The scope of the BMDP program includes:

     o   Univariate statistics (means, skewness) and

     o   Bivariate F and t tests and comparison of cells

     o   Scatter plots, probability plots

     o   Missing values estimates, patterns, and

     o   Multiple linear, stepwise, best subsets, principal
         component, and polynomial regression, as well as
         nonlinear regression

     o   Analysis of variance of fixed effect or mixed
         models, with repeated measure and unequal cell size

     o   Nonparametric statistical analysis

     o   Factor analysis, canonical correlation,
         discriminant analysis

     o   Cluster analysis on cases, variables, or blocks

NCC-IBM USER'S GUIDE                                     JAN 1988

The BMDP manual (see below) and its updates identify three
catalogued procedures for the use of BMDP =  BIMED, BHDPLKED, and
BIMEDT.  All procedures are installed with the same names at the
NCC and are used as documented in the manual.  However, BIMEDT
and BHDPLKED are invoked only when running with an extremely
large set of variables or when including user-FORTRAN data-
transgeneration coding.

The suggested reference manual is Biomedical Computer Program P-
Series .

BMD Communications is a publication that offers both usage notes
and application examples for the BMDP package.  It is available
free of charge from the Health Sciences Computing Facility,
AV-111 Health Sciences Building, UCLA, Los Angeles, California

See the on-line usage aid data set JUSD.BMDP.DATA for BMDP
documentation and sample BMDP programs.   Sections
contains information on accessing usage aid data sets.

7.3.2    Statistical Analysis System (SAS)

Statistical Analysis System is an integrated system for data
management and statistical analysis.  It combines statistics
routines, plotting, data manipulation, and report writing

SAS has a PL/1-like language which is used to transform or delete
variables, create new variables, create new data sets, merge data
sets, and print all or part of a data set.   SAS accepts data in
most forms and can handle many input devices .  The system also
performs general data manipulation and many statistical functions
including =

     o   Linear and nonlinear regression

     o   Analyses of variance and covariance

     o   Multivariate analysis of variance

     o   Matrix manipulation

     o   Multiple and partial correlation coefficients

     o   Spearman's and Kendall's correlation coefficients

     o   Frequency and cross-tabulation tables

     o   Assorted printed and plotted output.

NCC-IBM USER'S GUIDE                                     JAN  1988

The catalogued procedure SAS supplies JCL for most SAS functions.
The symbolic parameters OPTIONS, SORT, PRINT, GREGION, and ADABAS
have been included in these procedures.   OPTIONS supplies a
parameter list to the SAS control program.  SORT specifies the
number of cylinders to allocate for each sort work file and has a
default of 4.  PRINT can vary the class of printed output;
default is A. GREGION can be increased from the default of 1536K
for larger jobs as uith SAS PROC GLM. ADABAS specifies whether
production or development ADABAS files are to be accessible via
PROC EXTRA.  The default is production ADABAS files.  A SYSIN DD
JCL statement pointing to the SAS program which is to be executed
must be supplied.

SAS is a CIIST which allows users to execute SAS interactively in
ISO.  The keyword parameters OPTIONS, LIBRARY, and ADABAS can be
used to specify a SAS option string, a load module library
containing user defined procedures and functions, and whether
production or development ADABAS files are accessible via PROC
EXTRA.  Users must allocate their SAS data sets to be referenced
in an interactive SAS session with the TSO ALLOC command.  (This
can also be done with the SAS TSO statement which allows
execution of TSO commands without exiting from SAS.)

In SAS, users may obtain information on SAS procedures,
statements, formats, informats, functions and call routines by
using the HELP statement.  A brief description of how to use the
HELP statement can be obtained by entering the following:


Examples of SAS statement sets and JCL,  as well as catalogued
procedures and CLISTS, may be found in JUSD.SAS.DATA member
CONTENTS.  Samples of various SAS functions may be found in
7.3.3    Econometric Software Package C ESP)

The Econometric Software Package is a computer language for the
statistical analysis of time series by the ordinary least squares
method or the two-stage least squares method.  It is designed to
carry out the computational and data processing steps which occur
routinely in econometric research.  ESP is a revised and extended
version of an earlier system called TSP (Time Series Processor).

Input to ESP is made through the system's own free format
language.   ESP features include:

     o   Fitting and Extrapolation

     o   Matrix Operations

NCC-IBM USER'S GUIDE                                     JAN  1988

     o   Time Series Plots

     o   Statistical Measures

     o   Normalized Random Number Generator

     o   Scatter Diagrams

     o   Exponential Trend Lines

Users may access the package through the catalogued procedure

Further documentation concerning ESP and its uses is contained in
the Econometric Software Package User's Manual and on-line in the
partitioned data set JUSD.ESP.DATA.

7.3.4    International Mathematical £. Statistical Library  (IMSL)

The International Mathematical and Statistical Library is  an
extensive collection of mathematical, statistical, and function
subprogram written in FORTRAN and supported by IMSL.  In most
cases, the subroutines are available in both single and double
precision forms and support the vector feature hardware available
on the IBM 3090 processors.  With these subroutines, complex
computational routines can be constructed quickly and

Some of the subroutines and their groupings include =

     o   Categorized data analysis

     o   Eigenvalue analysis

     o   Forecasting

     o   Econometrics

     o   Analysis of experimental design data

     o   Basic statistics

     o   Time series analysis

     o   Generation and testing of random numbers

     o   Interpolation

     o   Approximation
                              7-1 1

NCC-IBM USER'S GUIDE                                     JAN  1988

     o   Curve fitting

     o   Linear algebraic equations

     o   Sampling

     o   Utility functions

     o   Vector and matrix arithmetic

     o   Linear programming

IMSL Version 10 is the currently supported IMSL library at the
NCC, and JUSD.IMSL.DATA is the on-line data set providing further
information on this product.  For users still exeucting IMSL
Version 9 code, an interface is available for translating Version
9 object code to Version 10 executable code.

7 . 3 . M . 1    Catalogued Procedures

The NCC provides and supports several catalogued procedures for
IMSL.  The IMSL procedures are as follows:

     PROC          Function

     IMSL1CLG       Compile, link, and go using Level  10
     IMSL1CISP      Compile, link, and go using the interface to the
                    Level 9 single precision library
     IMSL1CIDG      Compile, link, and go using the interface to the
                    Level 9 double precision library

7.3.M.2    Publications

Copies of IMSL user manuals can be ordered from IMSL at the
following address =

     IMSL Customer Relations
     2500 Park West Tower One
     2500 City West Blvd.
     Houston, Texas  77042-3020
     Telephone:  (713) 782-6060

7.3.5    Procedure Library-Mathematics (PL-MATH)

The Procedure Library-Mathematics is a set of basic computational
procedures for PL/1  language users.

Individual procedures or a combination of them can be  used for


NCC-IBM USER'S GUIDE                                     JAN 1988

the following general purposes =

     o   Elementary array manipulations

     o   Solution of linear equations

     o   Eigen analysis and related problems

     o   Polynomial operations

     o   Numerical quadrature

     o   Numerical differentiation

     o   Interpolation

     o   Approximation

     o   Smoothing

     o   Zeros and extrema of functions

     o   Linear programming

     o   Systems of ordinary differential equations with
         given initial values

     o   Convergence acceleration

     o   Transforms

     o   Other special functions

The PL-MATH routines are written in the PL/1CF) language and do
not include I/O statements.  Many routines provide the option to
handle double-precision computations.

When compiling a program containing CALL statements to PL-MATH
routines, the user must supply the JCL statement below:


The compiler parameter "CPARM=MACRO" must be supplied on the
compiler step EXEC statement.

Further information concerning PL-MATH routines is contained in
the Procedure Library-Mathematics Operations and Procedure
Library-Mathematics Program Description.

NCC-IBM USER'S GUIDE                                     JAN  1988


Most technical data and other information are best communicated
and displayed by pictures.   There are graphics facilities and
software available at NCC that can be used to display data.
Appendix J of the printed (yearly) version of this manual
contains samples of graphics output.

There are three types of NCC-supported graphics software;

     a.  Interactive Graphics.  Conversational graphics
         systems that produce several types of plots
         interactively at the terminal from ordinary
         English-liKe commands.

     b.  Application Programs.  These are special purpose
         programs designed to produce a specific type of
         plot.  They operate under the direction of user-
         provided control cards  and data.  No programming is

     a.  Subroutine Libraries.  When incorporated into a
         user-written main program, these subroutines
         relieve the user of programming many commonly used
         graphics functions as well as provide him direct
         control of the pen.

Documentation is available on-line for each graphics package.
These on-line usage aid data sets contain the information
necessary for their use on the NCC-IBM:

     a.   Introductory and usage notes.

     b.   Sample JCL and/or CLISTS.

     c.   Sample programs and/or test data.

     d.   Vendor documentation ordering information.

Each data set contains a member  CONTENTS describing the various
members and a member 1ISTPDS containing JCL to list the data set
in its entirety.  Methods of routing CalComp output to the NCC
and WIC plotters are also discussed in the usage aids for
CalComp-compatible software as well as in Section 5.4.4.  The on-
line usage aid data sets do not  pretend to replace the vendor

For interactive graphics without user programming, TELL-A-GRAF
and SAS/GRAPH are recommended.  Users with unique applications
needing to write their own program should investigate DISSPLA,
the best subroutine library available on the NCC-IBM.

NCC-IBM USER'S GUIDE                                     JAN  1988

7.M.1    TELL-A-GRAF

TELL-A-GRAF is a conversational graphics system which produces
quality graphs from ordinary English commands.  TELL-A-GRAF
output may take on one of five "personalities" -- line plot,  area
chart, bar plot, pie chart, or a page of text.  The program
produces graphics interactively when invoked from a terminal
capable of drawing vectors, although any type of terminal may be
used to enter commands to create a graphics output file.  A
separate post-processor program can then use the plot file to
produce graphics output on a plotting device.

Detailed information concerning TELL-A-GRAF usage on the NCC-IBM,
supported graphics devices, batch procedures, and User Manual
ordering information may be found in the on-line data set

CUECHART is an easy-to-use, user-friendly graphics package.  All
that is necessary to prepare a graph with CUECHART is for the
user to select any of 384 standard stencils pictured in the
CUECHART User's Guide, invoke CUECHART, and respond to simple,
"English" prompts to provide the information necessary to tailor
the general stencil to specific needs.   The actual output of
CUECHART is a file containing TELL-A-GRAF commands.  TELL-A-GRAF
is  subsequently invoked, the CUECHART file "included," and the
chart plotted on any of a variety of graphic output devices .

On-line documentation describing CUECHART (and TELL-A-GRAF) usage
on  the NCC-IBM is contained in the data set OUSD.CUECHART.DATA.

The DATA CONNECTION is an integrated portfolio of tools designed
to solve data access problems.  As the name suggests, the DATA
CONNECTION builds connecting links between TELL-A-GRAF and the
many sources of data used to produce meaningful, up-to-date
graphical presentations.  Its components are the following:

     a.  REPORT CONNECTION - Reads tabular data from printed
         reports stored in any format.

     b.  FILE CONNECTION - Reads, selects, analyzes and
         compares data from any data file with uniform
         record formats.

         directly with user-written applications programs.

NCC-IBM USER'S GUIDE                                     JAN  1988

     d.  DECISION SUPPORT CONNECTION - Edits, modifies, and
         mathematically manipulates data before it is
         converted to graphic output.

On-line documentation describing the DATA CONNECTION  and it
relationship to TELL-A-GRAF is contained in the on-line usage aid

TELLAPLAN EXPERT is a project planning and management tool.
Using TELLAPLAN's simple, English-like commands, the user can
accomplish the following =

     o  Build detailed plans for ongoing projects and events
        and control their progress.

     o  Report progress with a variety of Gantt charts,
        cost/resource charts, or printed reports.

     o  Show planned, actual, and estimated start and end
        dates, in addition to dependencies, critical paths,
        slack time, and milestones.

     o  Create sophisticated structure diagrams including
        work breakdown structures, arrow diagrams, and
        activity-on-node diagrams.

More information on using TELLAPLAN EXPERT on the NCC-IBM is in
the on-line usage aid data set JUSD.TELAPLAN.EXPERT.DATA.
7.4.5    InterOrg

InterOrg is a program that allows users to interactively create
organization charts.  The charts may be displayed on a variety of
graphics devices.  Users specify the box names, box titles,
optional comments, reporting relationships, plot titles,
footnotes, options for line style, colors, fonts, page layout,
etc.  Prompting mode reads all input from the keyboard, whereas
menu mode reads chart information from files saved during
previous InterOrg sessions.

InterOrg runs as a front-end to ISSCO's TELL-A-GRAF and the full
range of TELL-A-GRAF devices, formats, and capabilities are
available to the user.  InterOrg creates a TELL-A-GRAF 'include1
file which contains all of the data and commands necessary to
allow TELL-A-GRAF to plot the organization chart.

NCC-IBM USER'S GUIDE                                     JAN  1988

For more information, see the on-line documentation file
7 . M . 6    DISSPLA

DISSPLA is a library of subroutines which enable a user to
present data in graphical form.  The basic DISSPLA package,
together with its optional modules, includes the capability to
produce two- and three-dimensional graphs, geographic maps,
contour maps, and business charts, plus an extensive array of
alphabets and font styles.

Documentation may be obtained through any of the regional ISSCO
sales offices, or their corporate headquarters (telephone number
619-452-0170).  Manuals include the DISSPLA User's Manual, the
DISSPLA Pocket Guide, and DISSPLA First Facts.

More information on using DISSPLA on the NCC-IBM is in the on-
line usage aid data set JUSD.DISSPLA.DATA.
7.M.7    6EOMAP

GEOMAP is an interactive computer mapping program used for
displaying, and analyzing thematic coded data.  Users may select
from various GEOMAP formatted cartographic files, including
state, county and ZIP code boundaries.  Geographically based data
may be displayed for analysis and presentation under user
control.   GEOMAP reads and categorizes existing data files,
calculates new values, or accepts data directly from the user.

GEOMAP is menu-driven and allows the user to exercise control
over the characteristics of the map, including title(s), legends,
colors, etc.  The user may also choose to overlay other boundary
files on the map.

GEOMAP runs as a front-end to TELL-A-GRAF and the full range of
TELL-A-GRAF formats and capabilities is available to the user.
GEOMAP creates three input files that are passed to a modified
version of TELL-A-GRAF.  These files contain the selected
cartographic coordinates, boundary/symbol information and the
TELL-A-GRAF command required to produce the map.

Information concerning GEOMAP usage on the NCC-IBM system may be
found in the on-line data set JUSD.GEOMAP.DATA.
7.M.8    SAS/GRAPH

The Statistical Analysis System has a graphics feature that


NCC-IBM USER'S GUIDE                                     JAN  1988

allows for high quality plots.   Some SAS/GRAPH capabilities
include lines, bars, pies, block charts, contours, maps, THREE-D,
and text pages.  Refer to Section 7.3.2 for more information
about SAS.

On-line documentation describing SAS and SAS/GRAPH usage on the
NCC-IBM is contained in the data set JUSD.SAS.DATA.
7.4.9    ISSCO-GKS

ISSCO-GKS provides an implementation of a Graphical Kernel System
(GKS) library.  GKS consists of a set of basic functions for
graphics programming that has been defined by the American
National Standard.  It is a low-level graphics system for use in
creating portable applications that produce computer generated
two-dimensional pictures on vector or raster output devices.
ISSCO-GKS applications can be ported to any installation
supporting a GKS of the appropriate level in the programming
language used in the application.  ISSCO-GKS is at level 1b in
the GKS classification system and uses FORTRAN 77 as the

The on-line documentation data set JUSD.ISSCOGKS.DATA contains
further information on using ISSCO-GKS on the NCC-IBM.
7.M . 10    SURFACE II Graphics

SURFACE II, a graphics system for the creation of spatially
distributed data, is capable of producing graphic displays on the
data center's CalComp 1051 plotters.  The basic SURFACE II
display is a contour map, a plot of two coordinates on which
values of a third variable are defined by lines of equal value.
SURFACE II will also produce a perspective block diagram which is
a two-dimensional representation of a three-dimensional object.
A third type of diagram is a posting, or plotting, of symbols
representing observations at their appropriate X and Y

User commands implement the self-contained SURFACE II routines
that specify details such as which plot option to perform, how
large to make the display, and where to find the input data.

The on-line usage aid data set JUSD.SURFACE2.DATA contains
detailed information about SURFACE II's usage on the NCC-IBM,
ordering information for the SURFACE II User Guide, and a
jobstream that can be used to produce sample SURFACE II displays.

NCC-IBM USER'S GUIDE                                     JAN  1988

7.4. 11    Basic Tektronix Software

The Basic Tektronix software at NCC consists of two subroutine
libraries.  Tektronix terminals are the only output devices.
DISSPLA is a much more comprehensive subroutine library and its
use is recommended over this software.

The TEKTRONIX subroutine library consists of the Terminal Control
System (TCS) and Advanced Graphing II (AGII) routines.  Simple
histograms and X-Y charts may be created easily from TCS alone or
embellished with the addition of the AGII routines into the user-
written program.

The PREVIEW subroutine library contains routines that enable  the
user to view his CalComp output on a Tektronix terminal.  This
capability reduces program development/debugging time since
turnaround is minimized to the system response time (as opposed
to waiting 24 hours for plotter output).

The on-line usage aid data set JUSD.TEKTRONX.DATA contains local
Tektronix usage information.
7 . 4. 12   Tektronix Interactive Graphics Library

The Tektronix Interactive Graphics Library (IGL) is a collection
of SIGGRAPH CORE compatible routines providing graphics output on
a variety of Tektronix devices .

DISSPLA is a much more comprehensive subroutine library and its
use is recommended over this software.

The on-line usage aid data set JUSD.IGL.DATA contains information
necessary for IGL usage on the NCC-IBM.
7.4.13   CalComp Software

CalComp subroutines are available on the NCC-IBM.  This software
may be plotted on either of the NCC's 1051 plotters or output to
a Tektronix-compatible graphics terminal by a method called

The subroutines fall into six categories, distinguishable by
general application:  basic, general, scientific, drafting,
business, and curve fitting.

For example, the BASIC routines allow the user to move the pen,
change the color, and draw annotation, axes, and numbers.
Supplementing these functions, the GENERAL routines plot circles,
dashed lines, polygons, etc.  SCIENTIFIC routines perform

NCC-IBM USER'S GUIDE                                     JAN  1988

logarithmic and polar coordinate functions, etc.  A description
of the available CalComp routines, as well as details of
subroutine and plotter usage, is contained in the on-line usage
aid data set JUSD.CALCOMP.DATA .

DISSPIA is a much more comprehensive subroutine library and its
use is recommended over this software.
7 . M . 1 M   Cartographic Automatic Mapping

The Cartographic Automatic Mapping (CAM) program performs a
variety of cartographic functions using any one of 18 different
map projections.  User programming is not required.  Graphic
output may be obtained on either of the Data Center's CalComp
1051  plotters.

The World Data Bank I and II files are used with CAM in the
creation of a geographical base map with the plotting of
coastlir-es, islands, lakes, and rivers possible within the user-
defined map boundaries.  The program is not well-suited, however,
for overlaying user data and annotating the base map.

The on-line usage aid data set containing local CAM usage
information is JUSD.CAM.DATA.
7.4.15   POLYVRT

POLYVRT is an application program from Harvard that can be used
to extract and manipulate selected geographic coordinate data
from the COUNTYDIME and WORLD DATA BANK I files.  Polygon-
formatted, varying resolution output files may be created.  This
data may be plotted by POLYVRT, but the best use is to create
files for later processing by DISSPLA or other programs.

Detailed information is contained in the on-line usage aid data

NCC-IBM USER'S GUIDE                                     JAN  1988

7 . M .16   Device-Driver Subroutine Libraries

In addition to the CalComp and Tektronix subroutine libraries,
two other device-driver subroutine libraries are available for
incorporation into a user-written main program:

DEVICE                      LIBRARY         ON-LINE USAGE AID

ZETA Plotters           SYS2.ZETA.LOAD     JUSD.ZETA.DATA

(PTC 6 Controller)

Higher quality graphics may be output on each of these devices
and many more through the user-friendly TELL-A-GRAF package (See
Section 7.M.1).

7.M.17    Graphical Data Display Manager (GDDM)

IBM's Graphical Data Display Manager (GDDM) is a series of
programs for creating, showing, and storing pictures.  It
provides support for IBM (or IBM-compatible) devices:  3270
display terminals, printers, plotters, and scanners.  The device
independent graphics software available on the IBM 3090 (such as
SAS-GRAPH,  TELL-A-GRAF, DISSPLA, etc.) uses GDDM to provide
support for these devices.

GDDM includes several end-user utilities, most notably tha
Interactive Chart Utility (ICU).  It is an interactive, menu-
driven utility program for producing various types of charts,
including bar charts, histograms, tower charts, pie charts, and
X-Y plots.

GDDM also includes an application programming interface.
However, it is recommended that DISSPLA be used for application
programs for its higher level of functionality and device

More information on GDDM can be found in the data set

7.5.1    PANVALET Program Management and Security System

The PANVALET Program Management and Security System supports the
storage, retrieval, maintenance, backup security, and control of
all IBM and non-IBM source programs, object programs, job control
language, card-image data files, and print files.

NCC-IBM USER'S GUIDE                                     JAN  1988

Some features of the PANVALET System include:

     o   Maintaining a central library file for programs,
         job control streams, data files, and print files.

     o   Protecting production programs and data files
         against change or deletion

     o   Restricting retrieval and access to confidential
         data files

     o   Providing complete backup of all programs, job
         control streams, and data files between libraries

     o   Increasing programmer efficiency

     o   Increasing computer throughput

     o   Reporting all pertinent information and any
         activity on the library's contents

The PANVALET System consists of eight basic programs:  PAN#1,
PAN#2, PAN#3, PAN#M, PAN#6, PAN#7, PAN#8, and PCOMPARE.  PANtt! is
designed for general use by all programmers and is the most
frequently executed program.  PCOMPARE uill compare two data
sets.   The remaining programs are designed for use by systems
programming and operating personnel, as well as data processing
management.  These six programs provide the control and backup
security features of the PANVALET System.

A series of catalogued procedures are available to assist users
in executing PANVALET:

     o   PANCOBC - Selects a COBOL program from a PANVALET
         library and executes COBUC

     o   PANCOBCL - Selects a COBOL program from a PANVALET
         library and executes COBUCL

     o   PANINIT - Creates and initializes a PANVALET

     o   PANEXEC - Allows user to add, delete, update, and
         select members of PANVALET library list

     o   PANFTGC - Selects FORTRAN (G1) program from a
         PANVALET library and executes FTG1C

     o   PANCOMP - Compares two data sets.

Documentation for PANVALET may be obtained on-line in data set

NCC-IBM USER'S GUIDE                                     JAN  1988

JUSD.PANVALET.DATA.   To print the entire documentation data set
the member LISTPDS can be used.
7.5.2    Data Catalog 2 (DC2)

DC2 is a comprehensive data dictionary/directory system.   It may
be used to document data,  procedures, users, and their
interactions.  The system can be equally effective as a systems
development tool, as a data administration tool, and as an
inventory of user data and procedures resources.

To call DC2, users should execute the following PROC's:

         DC2BKUP (backs up DC2 files)
         DC2CNVRT (converts source programs to  dictionary  entries)
         DC2COPY (copies and moves between dictionaries)
         DC2INIT (initializes a dictionary)
         DC2MAINT (dictionary maintenance)
         DC2PURGE (deletes the master files)
         DC22RY  (queries the dictionary, generates COBOL layouts)
         DC2RPT  (report on the dictionary-no index report)
         DC2RPTI (report on the dictionary-index report)
         DC2RSTR (restores the dictionary)
         DC2UPD  (updates the dictionary)
         DC2UTL  (performs utility functions)

An on-line documentation data set (JUSD.DC2.DATA) has been
created to accompany the latest version of DC2.  This
documentation is intended to be all inclusive and will eliminate
the need for each user to acquire DC2 vendor documentation.  The
LISTPDS member of this data set should be used  to obtain a
complete copy.

CLIST's exist to aid in DC2 execution.  The CLIST's, their
functions, and the procedure to invoke them are listed in  member
CLISTS of the documentation data set.

OPTIMIZER III is designed to optimise COBOL programs and assist
the programmer in his testing, debugging, and source-level
tuning.  The OPTIMIZER III package consists of three components

The OPTIMIZER component produces a smaller and more efficient
program by eliminating redundant and unnecessary machine
instructions at the object code level.

NCC-IBM USER'S GUIDE                                     JAN  1988

The DETECTOR component is used to help debug COBOL programs.  It
produces three reports: Abend, Snap, and Intercept, which assist
the programmer in locating, analyzing, and correcting program

The ANALYZER component is used to assist the programmer in  the
testing/debugging and source-level tuning stages of program
development.  The ANALYZER reports tell the programmer which
sections of code are most time consuming, what sections are
inefficient, and what sections need further testing.  From  these
reports the programmer Knows where his time can be best used.

Since programs run faster with OPTIMIZER III, system resources
are available for other uses.  Therefore, OPTIMIZER III can lead
to improved throughput, better turnaround, and expanded peak
system capacity.

Use of OPTIMIZER III should be restricted to production level
programs as compile time triples.   It is not efficient to run in
a program-testing stage.
                                         i •
The current floor version load library is SYS2.OPTIII.LOAD.  The
OPTIMIZER III PROC's and their functions are as follows:

        OPT3C             Compile only.
        OPT3CL            Compile and link.
        OPT3CLG           Compile, link and execute.
        OPT3GO            Execute and generate DETECTOR reports.
        OPT3RPTS          Generate ANALYZER reports.

Documentation is available on-line in JUSD.OPTIII.DATA .   To print
the entire on-line documentation data set, the member LISTPDS may
be used.

7.7.1    General Purpose Simulation System (GPSS)

General Purpose Simulation System V, developed and supported by
IBM, is a program for conducting evaluations of and experiments
with systems, methods, processes, and designs.  A GPSS program is
built around a set of simple entities which are divided into four
classes:  dynamic, equipment, statistical, and operational.
Dynamic entities are called transactions and represent units of
traffic.  Equipment entities represent objects which are acted
upon by transactions.   Statistical entities are queues and tables
which measure system behavior.  The operational entities provide
the logic of a system, instructing the transactions where to go
and what to do.  Once a system is defined in this manner GPSS V
provides as output:

NCC-IBM USER'S GUIDE                                     JAN  1988

     o   The amount of transaction traffic flowing through
         the complete system and/or any of its parts

     o   The average time for transactions to pass through
         the complete system or between selected points, and
         the distribution probability of this passage time

     o   The degree to which each item of equipment in the
         system is loaded, together with the distribution of
         storage occupancy

     o   The maximum and average lengths of queues occurring
         at various points, as well as their distribution

The GPSS executable load modules reside in data set
SYS2.GPSS.LOAD.   Users can access the package using catalogued
procedure GPSSVGO.  This procedure provides the JCL to execute
all GPSS functions.  The symbolic parameters STORAGE, PRINT,  and
GPSSREG are provided to tailor the procedure.  STORAGE is a
parameter which is passed to GPSS.  PRINT governs the class of
the printed output; the default is A.  GPSSREG is the region  size
in K bytes  which GPSS is to operate in;  the default is  100K.
The GPSS V  User's Manual lists permissible values and their

7.7.2     Simulation PL/1 (SIMPL/1)

Simulation  Language for PL/1 is a PL/1-based simulation  language.

SIMPL/1 requires that a model of the system being studied be
coded  in the SIMPL/1 language.  This is then processed by the
SIMPL/1 preprocessor which produces PL/1 source code, which may
then be used as input to the PL/1 Optimizing or Checkout
Compiler.   As a rule, many PL/1 statements are generated from
each SIMPL/1 statement.

The SIMPL/1 executable load modules and library routines reside
in data sets SYS2.SIMPL1.LIB, SYS2.SIMPL1.SIMLIB1,
SYS2.SIMPL1.SIMLIB2, and SYS2.SIMPL1.DIAGFILE.   Users can access
the package with the following catalogued procedures:

     o   SIMXP    - preprocessor

     o   SIMXPC   - preprocessor and compile

     o   SIMXPCG  - preprocessor, compile and go

     o   SIMXPCLG - preprocessor, compile,  link-edit, and go.

NCC-IBJ1 USER'S GUIDE                                     JAN  1988

These procedures contain the JCL to invoke the SIMPL/1
preprocessor, pass the output to the PL/1 Optimizing Compiler,
and create and execute the resulting load module.  Symbolic
parameters can be supplied by the user to change the class of the
output, change the region size, and supply parameters to the
SIMPL/1 preprocessor.

Further information about SIMPL/1 is contained in the SIMPL/1
Reference Guide.

Version 3.7 of the SCRIPT system, a product of the University of
Waterloo, Ontario, Canada, is available to NCC-IBM users.  SCRIPT
is a text processing program that can be used to format files of
fixed or variable length or special format records.   Using
control commands, a page format can be established (line length,
number of lines per page, etc.) and headings and footers can be
specified.  Other commands provide the capability to justify
lines, indent blocks of text, incorporate footnotes, and build a
table of contents and an index.  The capabilities of SCRIPT are
many and varied, but it can be used effectively as a simple text
formatter with only a handful of commands.  Comprehensive
documentation, including the vendor-supplied manuals, usage tips,
and JCL to obtain listings or to execute SCRIPT, are available
on-line in data set JUSD.SCRIPT.DATA.   The executable load module
is in data set SYS2.SCRIPT.LOAD.

SCRIPT was used to prepare this User's Guide.

FOCUS is a comprehensive information processing system designed
for interactive or batch processing.  Current versions of FOCUS
are available on the IBM 3090 and Regional Logical Mainframes
(IBM 4381's).  FOCUS is also available for IBM XT and AT personal
computers with sufficient memory capabilities.

FOCUS contains facilities for describing external files and FOCUS
files and for generating reports.  FOCUS provides a dialogue
manager enabling incorporation of full-screen maintenance
procedures and validation of data based on user definitions.  The
system is structured so that it may be used by nonprogrammers as
well as programmers.  FOCUS is powerful enough for developers to
build 3270 menu-driven aplications yet easy enough for
nontechnical end users to perform their own ad hoc queries and

ISPF users can invoke FOCUS through the primary menu or by

NCC-IBM USER'S GUIDE                                     JAN 1988

entering FOCUS on the SPF command line.  Conventional terminal
users can invoke FOCUS by entering FOCUS under TSO.   FOCUS batch
jobs can be submitted from TSO.

Information about FOCUS, its use, documentation, and classroom
training is available on-line in JUSD.FOCUS.DATA.   A list of
available documentation is included in Section 7.9.7.
7.9.1    Capabilities of FOCUS

The capabilities of FOCUS may be viewed in terms of typical
information processing system requirements.  Such systems can be
divided into the following components =

     o  On-line data entry, editing, and updating of master
        data files as a single, continuous process.

     o  On-line data entry, editing, and creation of
        -transaction files, followed by batch updating of
        master data files, as two separate processes.

     o  Sub data file creation using extraction of records
        and/or fields from a production data file(s) and
        storing them in a data file structure for back-end
        processing.  Sub data file creation may involve
        transmission of files from mainframes to PC's, or
        vice versa, and transmission from mainframes to
        distributed logical mainframes (LMF's).

     o  Back-end processing which primarily consists of
        retrieval, presentation, and analysis of data
        maintained in the transaction files, master data
        files, or sub data files.  Back-end processing may
        involve canned (routine) or ad hoc activities.
        Canned processing is a standardized procedure which
        can be performed by executing a few commands.
        Typically, canned programs are repetitive or
        commonly used functions (such as standard, routine
        reports and queries).
7.9.2    Major Capabilities

FOCUS is an integrated information control system that provides a
nonprocedural English-like command language and a DBMS.  Its
three major capabilities are data retrieval and analysis, data
file structure and storage, and data maintenance.

NCC-IBM USER'S GUIDE                                     JAN  1988

The following capabilities are provided for retrieval and
analysis =

     a.  FOCUS and PC/FOCUS allow access to FOCUS or non-
         FOCUS files.  FOCUS can retrieve information from
         ADABAS, S2K, IMS, VSAM, ISAM, Sequential, and FOCUS
         files .

     b.  A simple set of English-like commands, in concert
         with master file descriptions, allows quick
         generation of tabular reports and graphs.

     c.  The menu-driven, dialog-oriented, interactive
         statistical programs provide a comprehensive
         statistical analysis capability.

     d.  The graphics modules allow creation of simple
         business graphics, such as bar charts, pie charts,
         and histograms.

FOCUS data files use a "shared relational structure" data model.
This model supports both simple and complex applications with
small or large volumes of data.  The data file facility also
provides dynamic file joining (cross-referencing) and logical
file views.  In addition, data encryption and security at the
field-value level are available for FOCUS files.

For data maintenance, FOCUS offers a full-transaction processing
facility including a data file editor, 3270 full-screen
processing, batch transaction processing, and central data file
control for multiuser access and processing.

7.9.3    Extended Capabilities

Information Builders, Inc., the developers of FOCUS, have
extended the major components of FOCUS to the IBM PC/XT- and
PC/AT-compatible personal computers.   The personal computer
version is compatible with the IBM mainframe version.  This
compatibility includes the capability to transfer application
code and sequential files and then have them executed on
mainframes, or vice versa.

Information Builders, Inc. has introduced the concept of "Talk"
technology.  "Talk" technology allows the most novice user to
become quickly productive through menu-driven questions that
appear on a full-screen terminal.  For example, they can use
TABLETALK to develop and execute reports, and they can use
FIIETALK to create data base file structures.

NCC-IBM USER'S GUIDE                                      JAN  1988

All the "Talk" technology operates on the IBM PC/XT,  PC/AT, and
IBM mainframe.
7 . 9 . M    Security Capabilities

FOCUS has complete Data Base Administrator facilities  that  allow
the user to specify security for data access and  updating.   READ,
WRITE, READ/WRITE, and UPDATE can be controlled for FOCUS files.
For external files, however, such control is limited to  READ
access only, because FOCUS does not maintain. non-FOCUS files.

FOCUS can protect data at four levels:

     a.  File.

     b.  Segment.

     c.  Field.

     d.  Value within field.

Conditional access can also be specified in the sense  that  access
will be allowed only if certain specified con.dition( s ) are  true.
A fifth level of security, data encryption, is available for
FOCUS data files only.  This level permits encryption  not only  of
the raw data but also of the Master File Description and
associated FOCUS routines.  The security provisions for  a file
are specified in the Master Dictionary entry which in  itself is

FOCUS can also limit the number of records read from any file as
well as control the number of retrieved records that may fulfill
IF screening tests in a user's request.

7.9.5    Advantages and Disadvantages

The following major advantages of FOCUS summarize its  value to
NDPD users:

     a.  Single data processing language interface.  As
         such, it provides data base access to S2K, ADABAS,
         fiSAM, ISAM, and VSAM files with the same report
         generation language syntax.

     b.  Compatible across three hardware levels.  These
         include the IBM 3090, the IBM M3xx-compatible
         logical mainframe, and the IBM PC/XT and PC/AT  and
         their compatibles.  Applications developed on any
         of these processors can be transferred with little

NCC-IBM USER'S GUIDE                                     JAN 1988

         code conversion.  This compatibility will reduce
         training costs and permit better use of personnel

     c.  Permits faster application development.  Once a
         famililarity with the product has been established,
         applications can usually be developed more quickly
         than with the use of procedural language.

     d.  Provides "Talk" technology, thereby allowing the
         novice user to become productive by eliminating the
         need for keyboard typing.  Because the user must
         only move a cursor to desired options described in
         simple English, a complete report request, for
         example, can be developed without knowledge of
         FOCUS commands.

FOCUS also has several disadvantages of which users should be

     a.  Like all Fourth Generation Languages, FOCUS can
         place huge  demands on computing resources.  Good
         file design and control of file size are therefore

     b.  FOCUS may not be suitable for complex statistical
         analyses.  Although FOCUS is good for simple
         statistics  and adequate for complex statistics, it
         is not designed to compete with SAS for complex
         statistical analysis.

     c.  FOCUS provides limited graphics, consisting of pie
         charts, histograms, and bar charts.  Although it is
         good for simple business graphics, its usefulness
         for more complex graphics is limited.

7.9.6    FOCUS Software Available on the IBM 3090

The following basic  system software is available:

     o  FOCUS Report Generator and Dialogue Manager for
        reporting from FOCUS and/or external files.

     o  FOCUS Data Base Management, Transaction Processor,
        and Interactive Data Base Editor.

NCC-IBM USER'S GUIDE                                     JAN  1988

Optional features include the following;

     o  FOCUS Graph Subsystem

     o  FIDEL (full-screen data entry language)

     o  Simultaneous Users

     o  FOCUS Statistical Analysis Package

     o  FOCUS Host Language Interface

     o  TABLETALK (window-driven process of creating FOCUS
        report requests)

     o  FILETALK (window-driven process of creating FOCUS
        report requests)

     o  FOCUS Data Dictionary

     o  PC Data Export

Optional data base interfaces are as follows:

     o  ADABAS Interface to Report from ADABAS Files

     o  System 2000 Interface to Report from System 2000

7.9.7    Documentation

A variety of technical documentation on FOCUS is available from
the software vendor.  All orders for documentation should be  sent
to the following address:

     Publications Order Department
     Information Builders, Inc.
     1250 Broadway
     New York, New York  10001

     Telephone Number:  (212) 563-9120

Following is a list of documentation:

     a.  General Reference Materials

         o  FOCUS Users Manual - Current Release

         o  FOCUS fiuick Reference Guide

NCC-IBM USER'S GUIDE                                     JAN  1988

     b.  Documentation for New Users

         o  FOCUS guery Language Primer

         o  An Introduction to FOCUS Through a Sample Application

         o  Getting Started in FOCUS

         o  TABLETALK and FILETALK Usage Instructions

         o  User Workbench

     c.  Data Dictionary Users Manual

     d.  Interface Users Manuals and Installation Guides

         o  ADABAS Interface Users Manual

         o  FOCSAM Interface Users Manual

         o  MVS/TSO Interface Users Manual

         o  System 2000 Interface Usars Manual

         o  TELI-A-GRAF Interface Users Manual

     e.  Additional User Aids

         o  FOCUS Techniques

         o  User-Written Subroutines

7.10    MEGACALC

MegaCalc is an electronic spreadsheet program for IBM mainframe
computers.  It is easy to use and includes on-line HELP
information.  The user must have full-screen (3270-type) access
in order to use MegaCalc.

Extensive facilities are provided with MegaCalc.  Data base
facilities are provided as well as the facility to process data
from data sets external to MegaCalc.  MegaCalc supports uploading
of VisiCalc, SuperCalc, and LOTUS 1-2-3 spreadsheets from a
personal computer.

To invoke MegaCalc from your TSO session, enter the following


NCC-IBM USER'S GUIDE                                     JAN 1988

A MegaCalc support group has been established.   Call (714)
474-0800 for more information.


                   JAN 1988
NCC Service Areas =   Office Telephone Numbers

Listed are the telephone numbers for the primary NCC-IBM  service
areas. For numbers associated with services other than  those
listed, contact the NCC User Support Department at  the  number
given below;
Central Data Base Administrator
Research Triangle Park, NC  27711

Data Processing Support Services
Research Triangle Park, NC  27711

Disk Librarian
FIMAS Office
Research Triangle Park, NC  27711

I/O Control Desk
Shift Supervisor (NCC)

NCC Communications and Training
NCC-IBM Production Control
     Neu Agreements
     Operations and Problems
NCC-IBM System Status Recording
(919)  541-2385
(FTS)  629-2385
(800)  334-9700
(919)  541-2385
(FTS)  629-2385
(800)  334-9700
541-1 137
629-1 137
(919)  541-2385
(FTS)  629-2385
(800)  334-9700

(919)  541-7862
(FTS)  629-7862
(800)  334-2405





             JAN 1988
NCC User Support
Research Triangle Park, NC  27711

NCC Network Control Facility
Research Triangle Park, NC  27711

Network Support Group
Tape Librarian
Research Triangle Park, NC  27711

TSSMS Office
Research Triangle Park, NC  27711

Washington Information Center
401 M Street, SW
Waterside Mall
Washington, DC  20024

WIC Operations Supervisor

WIC Production Services
           (919) 541-7862
           (FTS) 629-7862
           (800) 334-2405
           (919) 541-4506
           (FTS) 629-4506
           (800) 334-0741
           (919) 541-2385
           (FTS) 629-2385
           (800) 334-9700
           (202) 488-5900
           (800) 424-9067
           (202) 488-5980

           (202) 488-5970
NCC Key Personnel: Office Telephone Numbers

         Don Fulford
         Maureen Johnson
           (919) 541-3970
           (FTS) 629-3970
           (919) 541-3483
           (FTS) 629-3483

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN 1988

                    APPENDIX B=  SAMPLE FORMS

                     USER PROFILE WORKSHEET

Please enter/verify:


      PHONE:                           USER-ID:

Computer System:        	 IBM     	 SPERRY    	 DEC
                       	 IBM PC  	 PRIME     	 LEXITRON

NCC Publications:       	 NCC-IBM User's Guide
                       	 NCC-SPERRY User's Manual

Do you wish to receive ALL user memorandums, regardless of
subject area?      	 Yes      	 No

If you need only selected information, check the subject areas of
your choice:

           	  Data Center Management
           	  General Information
           	  Data Center Policy, Usage
           	  Telecommunications System
           	  Technical Advisories
           	  Software (All Software)
           	     Systems Software
           	     Compilers (All Compilers)
           	     Application Packages (All Packages)
           	        Database Management Systems
           	        Retrieval/Reporting Systems
           	        MATH/STAT Systems
           	        Project Management Systems
           	        Plotting/Graphics Systems
           	        File Management Systems
RETURN COMPLETED FORM TO ••    Communications and Training
                            Research Triangle Park, NC  27711
                            Attn:  User Profile

NCC-IBI1 USER'S GUIDE       APPENDIX B                    JAN 1988

                          SAMPLE FORMS

Please use this form to express your opinions and comments
concerning this publication.  We are particularly interested in
your views concerning the completeness, technical accuracy, and
Please complete the following;

     o  Commentator's Name:  _

     o  Telephone Number:    _

     o  Title:
     o  How do you use this publication?
                   Frequently for reference
                   For introductory purposes to a subject
                   For information on a specific topic
                   Other (please specify):
        Comments (Please include page numbers and give
     o  Do you require a response?
This form, together with any supporting material, should be
forwarded to:

               NCC-IBM User Support
               Research Triangle Park, NC  27711
Thank you for your comments.

NCC-IBM USER'S  GUIDE        APPENDIX B                       JAN  1988

                          REQUEST  FOR REFUND


                           USER  INFORMATION


PHONE  	   MAIL  CODE 	  REGIONS    1-10
                                                        HDQTRS       11
STREET	  LAB           12
                                                        CONTRACTOR   13
CITY,  STATE,  ZIP 	  OTHER        14


     JOB                            START   END
     NAME       JOB #      DATE     TIME    TIME   AMT    ACCT

1 .  	  	     		

2 .  	  	     		



1 .  	
2 .

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN  1988

Pages B-M through B-12 are not on-line.  Once a year the NCC
prints the NCC-IBM User's Guide and sends a copy to each user
registered on the User Profile.  These pages are included in  the
printed version.

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN 1988


The NDPD Telecommunications Service Request (TSR) consists of
six pages containing all anticipated user needs for NDPD
telecommunications service.  The TSR addresses
telecommunications support services only, not procurement of
ADP equipment.

Each page of the TSR is accompanied by instructions for completing
the form page.  Although the TSR is largely self-explanatory, the
instructions provide additional clarification regarding certain items

The TSR contains the following six pages ••
     PAGE 1
     PAGE 2
     PAGE 3
     PAGE M
     PAGE 5
     PAGE 6
NDPD Telecommunications Service Request
Full-Screen Application Terminal Equipment
Graphics and ASCII (TTY) Terminal Equipment
RJE Registration Form
Projected Equipment Activity
Terminal Activity by Function and Host Systems
PAGE 1  of the TSR must be included with every request.  Otherwise,
only those pages necessary to document the request should be
submitted.  Note that page 1 contains a narrative section to allow
users to explain specifically what is needed.

Please  fill out the requested information as thoroughly as you
can.  Upon receipt of the form by NCC, our telecommunications staff
will work with you to supply any missing information.

     Return the NDPD Telecommunications Service Request to ••

     RTF, NC  27711


               JAN 1988

                PAGE 1
ADP Coordinator Name

     Contact Site Location




     City 	State
     Requestor or Contact Person
     Commercial Phone 	
EPA Project Officer
  Room Number
  Zip Code
  FTS Phone

  Phone (	

                        Requested Completion  Date
This change will be=   Permanent
If Temporary, what is duration?  Months

    RTP, NC  27711
  i	1
  I  TAG RECEIPT:      |
  I  TSR No.	  |
  I  Date 	  |
  I  Time  	  |
  I  Title	  |
  L	1

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN 1988

PAGE 1  of the TSR must be submitted with every request.  Note
especially the following items:

Signature:  The signature line must be signed only by the
    ADP Coordinator.  The signature represents approval of the
    request for service.

Site Location ••  Provide the exact location at which the
    service/equipment is to be received.  Note that the room number
    and the building should be included.

Contact Person:   Provide the name and telephone numbers of
    the person who can specify exactly the parameters surrounding
    the request.

EPA Project Officer:  If the service/equipment requested is
    for a contractor site, the name of the EPA Project Officer must
    be  listed.
Narrative Explanation =   Describe your request as fully as
    possible.  Explain any information that you believe needs
    clarification or justification.
Permanent or Temporary Change•   If the request is for
    temporary service/equipment, specify the length of time for
    which the service/equipment will be needed.

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN  1988

                            Terminals      Printers     Controllers



Existing Full-Screen Equipment:

    Terminals 	   Printers




                   Protocol Converter Ports
Existing Full-Screen Equipment:  Terminals 	


From:  Building 	  To:  Building









NCC-IBM USER'S GUIDE       APPENDIX B                    JAN  1988

                     INSTRUCTIONS FOR PAGE 2
PAGE 2 of the TSR should be submitted as necessary to document the
request.   It must be submitted if the request is for "true" full-
screen equipment or for full-screen-emulation equipment.  Full-
screen equipment includes the IBM 3270, the SPERRY Uniscope, and
other terminals.  Completed samples of this page are available
from the  NCC/NETMORK SUPPORT GROUP or in the data set NCFJCMT.TSR.


                              Terminals   Printers



                                JAN  1988

                                PAGE 3
                               Host Type



                              Terminals   Printers       Host  Type

From:  Building 	





           To--  Building





State 	 Zip

NCC-IBM USER'S GUIDE       APPENDIX  B                     JAN 1988

                     INSTRUCTIONS  FOR  PAGE  3.

PAGE 3 of the TSR should be submitted  as  necessary  to  document the
request.  Completed samples are provided  upon  request  by the
NCC/NETWORK SUPPORT GROUP.  Note that  page  3 applies  only to graphics
terminals and ASCII TTY's.


                      RJE REGISTRATION FORM
                 OAN 1988

                  PAGE 4

Terminal Manufacturer 	

Terminal Model Number 	

Line Protocol (BSC/SDLC)

Line Printer Speed 	
Existing Line Type

Host Access 	

Number of Readers 	

Number of Printers 	

Number of Punches 	
 Protocol Equivalent

2770, 3781        	



S360/20 Mod 4,5   	

S360/20 Mod 5,6
S360/22, 25, etc.

1 130

System 3



                                        i	i

PRINTER SETUP           PR1             PR2            PR3

Print Positions          	           	           	
Separator PAGE           	           	           	
Drain Start              	           	           	

                                                  i	1
     Compress/Expand Feature 	 	      I  FOR NDPD USE ONLY I
     Horizontal Format Feature 	 	      I   RMT ID =          I
     Additional Buffer Expansions	 	      I    IBM    	I
     Multiple Record Features 	 	      I    SPERRY 	I
     Transparency 	 	      I    p/w    	I
                                                  i	i
     Buffer Expansion 	 	
     Console Support 	 	

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN  1988

                     INSTRUCTIONS FOR PAGE M.

PAGE 4 of the TSR should be submitted as necessary to document the

Terminal Type

Provide the following information:  manufacturer; model number;
communications line protocol to be used; line printer speed;
existing line type (either dedicated-circuit communications or
dial-up communications); and the CPU host(s) to be accessed (IBM,
SPERRY, PDF, etc.).

Terminal Peripherals

As applicable, list the number of readers, printers, and punches.

Printer Setup

Enter the special characteristics for each of your printers.

  Print Positions:  Enter the number of horizontal print positions
       for each printer.  The default is 132.

  Lowercase Support:  Indicate if the printer supports uppercase
       and lowercase printing.  The default is uppercase.

  Separator PAGE:  Indicate if you wish to have the standard  JES2
       separator page printed with each job printed.  The default
       is to print separator pages.

  Drain/Start:  Indicate if you wish to have the drain at RJE

Terminal Features

Make a checkmark on the blank line beside the desired categories.

Protocol Equivalent

Make a checkmark on the blank line beside the type of RJE station
the site is emulating.   Only one box should be checked.



                                        JAN 1988

                                          PAGE 5
I Receiving I From I To I File
Application I Device I Location I Location I Size
1 Freq.
Application 1
Average 1 Peak I Hours I
Lines/Day I Lines/Day 1 Per Day I CPU
1 ! 1
1 1 1
1 1 1
 Application  I
  From    I     To    I  File I        I   Time
Location  I  Location !  Size I  Freq. I  of Day

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN  1988

                     INSTRUCTIONS FOR PAGE 1
PAGE 5 of the TSR should be submitted as necessary to document the
request.  It must be completed if new applications or equipment
dictate new requirements .

File Transfer ( Personal Computer )

Since file transfer can affect telecommunications requirements
throughout the network, this section must be completed if the user
anticipates file transfer on a regular basis to support a particular
application .

    Application:  Specify the applicaton name with which the data
        is associated .

    Receiving Device:  Identify the device receiving the file (IBM
        PC/XT, IBM 3081, etc.).

    From Location:  Specify the computer system storing the original
        file that will be transferred (IBM PC/XT, IBM 3081, etc.

    To Location:  Specify the computer system that will receive
        the transferred file.

    File Size:  Specify the maximum number of megabytes of
        data that will be transferred at one time (one byte
        per character )  .

    Frequency;  Specify the frequency of the file transfer (daily,
        weekly, or monthly) and, if possible, the time of day the
        transfer will occur.

Remote Job Printing ( RJE )

If the equipment prints in the unattended mode, these print hours
should also be included when specifying the hours per day.  Remember
too the importance of distinguishing between CPU's when providing
print information.

Bulk Data Transfer (Host to Host)

For guidance in completing this section, see the instructions above
for File Transfer.  Specify the time of day the transfer will occur.

(With the installation of  IBM MSXX's at different sites, software
will be available for automatically transferring files between
processors as part of a standard update cycle.  This capability
will also enable users  to  transfer files to another CPU for further
processing . )

              APPENDIX B
          JAN 1988

        PAGE 6
                                   HOST SYSTEMS
No. of Terminals
Avg . Input
Char/Trans ....
Avg . Output
Char/Trans ....
Avg . No . of
Trans/Hour ....
Avg . No . of
Trans . Peak
No. of Terminals
Usage (L, A, H)*
No. of Terminals
Usage (L, A, H)*
No. of Terminals
Usage (I, A, H)*
No. Char. Pntrs
Average Lines/Hr
Avg Char/Prt








*Usage Key:
L (Light Terminal Activity)   = <
A (Average Terminal Activity) = >
H (Heavy Terminal Activity)   = >
1  Hour Per Day
1  Hour Per Day
3  Hours Per Day

NCC-IBM USER'S GUIDE       APPENDIX B                    JAN  1988

                     INSTRUCTIONS FOR PAGE 6.

PAGE 6 of the TSR should be submitted as necessary to document the
request.   It must be completed if an additional terminal is added
to the netuork.  The data traffic information it provides will be
used in determining if the telecommunications equipment must  be
upgraded to provide sufficient service.  Since follou-up questions
will normally be required between the telecommunications analyst
and the requestor, the information provided on page 6 is considered
to be simply a starting point for further discussions.

The page is broken down by function across various systems or system
Host Systems

The system interface columns (TSO, SPF, and CICS) are divided into
local and RTP columns.  "Local" refers to the IBM processor located
at your site. "RTP" refers to the IBM processor at Research Triangle
Application Function

Because "function" affects the service required, categorize the
terminal activity by application function.  For the "hours per day"
give the average for the number of terminals specified for that
particular function.

     Data Entry:  Data entry information related to the transaction
         size and number of transactions will be available
         from the National System Manager and is particularly
         relevant for CICS activity.  It is not expected that
         data entry on the IBM systems will be made under TSO
         or SPF.  The only exception will be the use of TSO with
         the Fourth Generation Information Center software.

     Character Printer Output:  This section refers to an IBM 3287
         or other cluster-attached printer. In the DEC environment,.
         this unit is normally a terminal-auxiliary printer.

NCC-IBM USER'S GUIDE       APPENDIX C                    JAN  1988

Following is a list of the JES2 Control Statements available at
the NCC-IBM, a brief description of each statement's function,
and the specific form that each statement must take.  Note that
the statements must begin in column 1, but are generally free-
form after that.

For some of these statements there is an equivalent command or
operand under TSO and WYIBUR to accomplish the same function.
See the OS/VS2 TSO Command Language Reference Manual and the
WYLBUR Guide for details.   Additional reference may be made to
the MVS JCL Reference and MVS JCL User's Guide.

The statements are presented in alphabetical order.  For ease of
use, the following summarization by function is provided:

         Statements to control connection to the CPU:

         Statements for modifying JOB statements =

         Statements controlling job selection:
             CNTL (EXCLUSIVE or SHARED)

         Statements to control routing of output;

         Statements for communication with the operator:

AFTER Statement:  The AFTER statement is used to delay the
execution of the job with which it is associated until after the
execution of the job it specifies has been completed.  The form
of the AFTER statement is=

         col 1
             /*AFTER  jobname

where jobname is the eight-character job name of the job whose
completion this job must await.  The form of a job name on the
NCC-IBM is iiixx, where iii is a validated User-ID and xx is a
two-character job identification assigned by the user.

NCC-IBH USER'S GUIDE       APPENDIX C                    JAN  1988

Assume Job 1  specifies Job 2 with an AFTER statement.  If Job 2
is not in the system when Job 1 is initiated, Job  1 will start
processing.  A job may also include one and up to  three CNTL
statements not to exceed four AFTER, BEFORE, and CNTL statements
(see Combinations).

Another version of the AFTER statement is as follows:

         col  1
             /*AFTER hhmm

where hhmm is the time of day (military time).  The job will not
execute until after the time of day has passed.

This version  of the AFTER statement is used to delay the
execution of  the job until after the specified time of day.
However, if the time on on the AFTER statement effect a delay of
more that 20  hours, the AFTER statement has no effect.

BEFORE Statement:  The BEFORE statement is used to ensure that
the job in which it is included is execux_d before the job it
specifies.  Its form is;

         col  1
             /^BEFORE jobname

where jobname is the eight-character job whose execution this job
must precede.  At the NCC, the form of a job name  is iiixx, where
iii is a validated User-ID and xx is a one- or two-character job
identification assigned by the user.

The BEFORE statement has no effect in the following cases:

     o   The  job specified in the BEFORE statement is not in
         the  system when the other job is initiated.

     o   The  job specified in the BEFORE statement is
         already executing when the job containing the
         BEFORE statement is submitted.

CNTL Statement:  The EXCLUSIVE Control statement is used in
conjunction with the SHARED Control statement to control
scheduling of jobs which utilize the same resources.  The first
use of either a SHARED Control or an EXCLUSIVE Control statement
with a particular resource name establishes that name in a list
of such names maintained by JES2.   The name will be kept in the
list until all jobs which specify it have been completed.  While
such a resource name is in the list, any number of jobs which
specify that  name as SHARED may be concurrently selected for
execution.  However, a job which specifies that name as EXCLUSIVE
will be selected only when no other jobs specifying the same name

           APPENDIX C
JAN 1988
are executing, and once such a job is selected no other jobs
specifying that name will be selected until it has been

Note that the name for a resource or group of resources is
entirely unrelated to the real identity of those resources.  It
may be chosen arbitrarily by the resource's users.  The resource
names in the list are not associated with any particular User-ID,
so different users may specify the same resource by using  the
same resource name in SHARED Control and EXCLUSIVE Control
statements.   In order to eliminate unintended contention for
different resources, the convention of using one's User-ID as the
first three characters of all resource names has been
established.  For resources used by more than one user, an
agreement on the User-ID to be used must be reached.  The  form of
the Control statement is =
         col 1

where resource is a three- to eight-character resource name which
begins with the User-ID of the owner of the resource.

JES2 allows any combinations or multiples of /*AFTER, /^BEFORE,
and /*CNTL control statements, not to exceed a total of four
statements within a job.  Within a job, the statements may occur
in any order.
Combination 1.
Combination 2.
Combination 3.
Combination M.



                                       (disp = SHR or EXC)

NCC-IBM USER'S GUIDE       APPENDIX C                    JAN  1988

Combination 5.  ^*AFTER JOBNAME
                /*AFTER JOBNAME2
                /*BEFORE JOBNAME4
                       RESOURCE, disp
Combination 6 .   /*BEFORE JOBNAME
                /^BEFORE JOBNAME2
                /^BEFORE JOBNAME3
Combination 1  allows for 0-2 /*CNTL cards.  Combinations 2 and 3
allows for 0-3 /*CNTL cards while Combination 4 will accept 1-4
/*CNTL cards.

Note that specifying a SHARED or EXCLUSIVE attribute for a
resource has no effect whatever on the interpretation of the OS
disposition parameter.  Appropriate disposition parameters must
still be supplied for all data sets.  The JES2 statements allow
or inhibit selection of jobs for execution.  Once selected, OS
still enforces the exclusive or shared use of files as specified
in the DISP parameter of the DD statements.  By use of the SHARED
Control and EXCLUSIVE Control statements where appropriate, users
can prevent the wasteful situation in which a job has been
selected, and  is therefore "using up" an initiator, but can
perform no useful work because of a data set conflict with
another job .

JOBPARM Statement:  The JOBPARM statement is usad to supply
frequently-changed parameters to a job, or to override those
parameters in  the JOB card of the job in which it is included.
The form of the JOBPARM statement is =

         col 1
             /*JOBPARM operand-string

where operand-string is one or more of the following operands,
separated by commas.  The valid operands are =

    CARDS=nnnn     where nnnn is the number of cards to be punched
                   in hundreds (from 0 to 9999)

    COPIES=nn       where nn specifies the number of copies to be
                   printed of the job related output (from 1 to
                   30 copies)

    FORMS=xxxx     where xxxx is a valid form number to be used
                   for printed or punched output (see Section
                   M.S. 6 for the form numbers available)

    LINECT=nnn     where nnn specifies the maximum number of
                   lines to be printed on each page (from 0 to

        APPENDIX C
JAN 1988
                   255 lines)
where nnnn specifies the estimated number of
lines to be printed, in thousands (from 0 to
    BURST=Y(or N)  where the default burst characteristics of
                   all SYSOUT data sets produced by this job.
                   This characteristic is used to either burst
                   the data set (Y), or to continuously fold
                   the data set (N), when the data set is
                   directed to an IBM 3800 printer with the
                   burst option installed.

    RESTART        if this job is executing before a re-IPL
                   and JES2 warm start and cannot be
                   restarted from a step or checkpoint, JES2
                   does one of the following:

                   Y indicates that the job is queued for
                   re-execution from the beginning of
                   the job.

                   N indicates that no special action is to
                   be taken.

                   If RESTART is not specified, N is assumed.
                   However,  the installation may override
                   this default by JES2 initialization

    ROOM=xxxx      an alphanumeric value indicating a programmer's
                   room number to be placed on the job's separators
                   for routing SYSOUT data sets back to the
                   programmer (from 1 to 4 characters).

If not overridden by a JOBPARM statement or specified on a JOB
card, the following defaults will be in effect:  CARDS=20
(2,000), COPIES=1, FORMS=6111,  LINECT=61, LINES=20 (20,000),
RESTART, and a full JES2 System Log.

MESSAGE Statement:  The MESSAGE statement is used to display a
message to the system operator on the operator's console.  If it
is included in a job, the message will be prefixed with the  job
number.  If a MESSAGE card is not included within a job, the
message will be prefixed with the input device name.  The form of
the MESSAGE statement is:

         col 1
             /^MESSAGE message text

NCC-IBM USER'S GUIDE       APPENDIX C                    JAN  1988

where message text is any desired message up to 60 characters in
length (columns 72-80 of a MESSAGE card are ignored).  The
message text need not be bounded by quotation marks or any other

Note that at the NCC, there is no need to send messages to the
system operator concerning mountable devices, even when a write
enable ring is required in a tape reel.  The use of MESSAGE
statements  should be restricted to those occasions in which
communication with the operator is required due to a requirement
for special handling.

NOTIFY Statement:  The NOTIFY statement causes a job's
notification messages to be directed to the User-ID specified on
the NOTIFY  statement.  Print, punch, and execution routings are
not affected.  Use of the NOTIFY statement overrides any NOTIFY
specification on the JOB statement.

The form of the NOTIFY statement is:

     col 1         10
         /^NOTIFY userid

A user identifier (i.e.. Ill) causes notification messages to be
sent to the user whose User-ID is specified.

NOTE:  The  /*PRINT and /*PUNCH statements are currently available
at NCC, but they will be dropped with the planned software
upgrade.  Their function can be duplicated with the use of the
TYPRUN=COPY keyword on the JOB statement (see Section 3.7.1).

OUTPUT Statement:  The OUTPUT statement specifies characteristics
and/or options of a specific SYSOUT data set or group of SYSOUT
data sets.

The form of the OUTPUT statement is:

     col 1            10
         /^OUTPUT   code parameters

Where code  refers to all SYSOUT data sets within the job whose
code in the form number subparameter of the SYSOUT parameter
matches the "code" specified on the OUTPUT statement (1-4
characters).  Specifying code as "*" indicates that this OUTPUT
statement is a continuation of the previous OUTPUT statement.
This code should not be one of the NCC-IBM form numbers.

The parameters are:

     BURST=Y (or N)       Y indicates that the printed output
                         from a 3800 printer is to be burst

JAN 1988
                         into separate sheets.   N indicates
                         that the printed output is to be in
                         continuous, fanfold mode.  N is the
The name of a character arrangement
table for a 380C printer.  Each name
is 1  to 4 alphanumeric or national
characters; from one to four names
can be coded.  To specify more than
one name, use CHARS=(xxxx,xxxx. . . ) .

A value that specifies the maximum
number of lines or cards contained
in a logical page.  The default is
the value specified in the JES2
initialization parameter for the
device.  For printers, it can be
from 0 to 32767; for punches it can
be from 1 to 32767.  Note that if
this value is specified by the user,
the value may be too high,  and
backspacing becomes difficult.

A value that specifies the number
of logical pages to be printed before
the next checkpoint is taken.  The
default is the value specified in the
JES2 initialization parameter for the
device.  May be from 1 to 32767.
Note that if this value is specified by
the user, the value may be too high, and
backspacing becomes difficult.

A value indicating the number of copies
of printed job-related output that is
to be produced (from 1 to 255 copies).
If coded COPIES=000, JES2 will
default the copy count to 1.

JAN 1988
A value that specifies how many copies
of each page of the printed output are
to be grouped together.   When COPYG
is specified, it determines the number
of copies to be printed (that is, if
COPYG is coded, COPIES is ignored).  The
total number of copies printed equals
the sum of the specified group values.
The sum of the group values cannot be
greater than 255.  To specify more than
one group value, code COPY=(nnn,nnn...).
The maximum number of groups that can be
specified is 8.  This parameter applies
only to the IBM 3800 Printer.

One to four different destinations can be
specified for each output data set.  To
specify more than one destination, code
DEST=(xxxx,xxxx) where xxxx is an alpha-
numeric value iK".icating one of the
following =
                         Where N designates a node and R designates
                         a remote work station that is connected
                         to the node.  nn, nnn, or nnnn is 1-99,
                         1-999, or 1-1000 and mm, mmm, or in in mm
                         is 1-99, 1-999, or 1-1000.  The maximum
                         number of digits (n and m combined)
                         cannot exceed six.  A specification of
                         RO is equivalent to the specification
                         LOCAL at node N.
An alphanumeric value indicating the data
set forms control or carriage specifications
(from 1-M characters).

The name (1-4 alphanumeric or national
characters) of the forms overlay frame that
the operator is to insert into the 3800
printer before printing begins.

A value (between 0 and 255) that indicates
the number of copies to be flashed with
the overlay, beginning with the first copy
printed.  For the 3800 printer,  if FLASH
is specified and FLASKC is omitted, all
copies are flashed.




JAN 1988
An alphanumeric value indicating the print
and punch form (1-4 characters).

A value indicating the data set indexing
print position offset (to the right) for the
32 1 1  printer ( 1-31 ) .

A value indicating the data set indexing
print position offset (to the left) for the
3211  printer (1-31).
     MODIFY=module-name  The name (1-4 alphanumeric or national
                         characters) of a copy modification module
                         previously stored in SYS 1 .IMAGELIB that is
                         used to replace variable  data in the  printed
                         data set of the 3800 printer.
The table reference character (0-3) that
identifies a character arrangement table
specified on the CHARS parameter.

An alphanumeric value indicating the universal
character set specification (1-4 characters).
For more information the user should refer to the OS/VS2 MVS JCL

The OUTPUT JCL statement is also used to specify processing
options  for a SYSOUT data set.  Refer to Section 4.1.4 of this
manual for more details.

ROUTE Statement:  The ROUTE statement is used to route printed or
punched  output (SYSOUT = A or SYSOUT = B) to s. specific location.  To
route printed output to a NCC printer, no ROUTE statement is
required although you may enter =

         col 1
             /*ROUTE PRINT LOCAL

and to route printed output to remote nnn, enter:

         col 1
             /*ROUTE PRINT RMTnnn

Note that the alternate forms of the operand are Rnnn, RMnnn,
NnRnn, or Nn, beginning on or before column  16.  Leading zeroes
should not be included in any of the forms of this operand.  To
route punched output, the card format is exactly the same as
above except that the word "PRINT" is replaced with the word
"PUNCH".  The word REMOTE (spelled out) is not valid in ROUTE

NCC-IBM USER'S GUIDE       APPENDIX C                    JAN  1988

The location can be a particular remote as indicated by Rnnn,  or
if the oputput is to be directed to the NCC-IBM from an LMF,  you
must specify N1 for Node 1.  Each node can have remotes
associated with it such as N1R255.  The LMF system is a node  in
the EPA IBM SNA network.  Therefore, to get a printout that is
executing on the NCC-IBM and have it print in New York, Region
II, the following is used:


SIGNOFF Statement:  The SIGNOFF statement is used to disconnect
an RJE station from the NCC-IBM computer system.  Its form is:

         col 1

SIGNON Statement:  The SIGNON statement is used to connect an RJE
station to the NCC computer, and to establish the remote number
for that station.  The User-ID and "RJE password" are required to
guard against unauthorized use of the NCC computer facility.
(Note that the "RJE password" is not the same as the general  NCC
password referred to elsewhere in this manual — it does not
satisfy the requirement [of the Security System] for confirmed
user identification, and may not be changed by the user.)  The
form of the SIGNON statement is:

         col 1               16
             /*SIGNON        REMOTEnnn

where nnn is the remote number assigned to the RJE station being
connected (entered without leading zeroes).

NCC-IBH USER'S GUIDE       APPENDIX D                    JAN  1988


The JES2 Operator Commands are the commands used by the operator
of an RJE station to control its devices and the jobs whose
output is routed to them.  The commands documented here are the
most commonly used standard JES2 commands.  Usually these
commands are entered at a console device supplied with the RJE
station, but some stations use a card reader for this purpose.
In any case, some action must be taken by the operator, prior to
command entry, to indicate that what follows is a JES2 command.
Refer to the manuals supplied with the RJE station to determine
the precise form of this action.

Operators should note that with JES2, the capability exists to
eliminate the printing of banner pages between output data set
groups.  This can be done by issuing a '$TRnnn.PRn,S=N' command.
The change can be made permanent by calling NCC
Telecommunications Support (see Appendix A for telephone number)
and requesting a JESPARM change to reflect this status.

Following is a brief description of the commands supported at the
NCC.  For convenience, the commands are grouped into those
concerning the job queue, those concerning job status, those
concerning device status, and a few miscellaneous commands.

The following words, when used in command format specifications,
will have the special meanings indicated.

operand - indicates the specification of all the jobs
          queued for a particular action or resource.
          The values are =

          fi=XE2,IBM -  to indicate jobs awaiting execution

          2=XE2x,IBM  - where x is a valid job class, to indicate
                        jobs awaiting execution in class x

          fi=PPU,IBM -  to indicate jobs awaiting print or punch
                       output processing

          2=HOLD,IBM - to indicate jobs waiting for any activity and
                       in hold status

          2=OUT,IBM  - to indicate jobs awaiting output processing

               APPENDIX D
GAN 1988
     Note:   With JES2 2.1.5 and the Mellon Bank mods, $DN and $D2
     command operand formats changed.  Unless the commands are
     followed by ",IBM",  the syntax may fail.  Examples follows:
         *DN,e=XE2A,IBM or $DN,XE2A
          NCC JES2:


          NCC JES2:
joblist - indicates the specification of one to five
          jobs, or ranges of jobs, by their JES2 job
          number.   Its form is one to five entries,
          separated by commas, where each entry is
          either a single job number or a pair of job
          numbers  separated by a hyphen (-) to indicate
          a job range.  If, in the entry of a range,
          the second job number is not greater than the
          first, only the second job number will be
          used by  JES2.
1jobname'  -
indicates the specification of a job by its
job name, enclosed in single quotation marks.
At the NCC,  a job name always consists of a
six-character User-ID followed by a one- or two-
character job identification suffix.
remote-number - indicates the specification of a remote
                workstation (JES2 RJE station).  Its form is


                where nnn is a one- to three-digit remote
                number, entered without leading zeroes.

device - indicates the specification of one to five
         devices,  separated by commas.  Each device is
         specified as:


         where Rnnn is the remote-number, entered as
         outlined  above; xx is "PR" (to specify a
         printer), "PU" (to specify a punch), or "RD"
         (to specify a reader); and i is the device
         number (there may be more than one of any
         type).  Stations which have only one of a
         particular device should enter "1" as the
         value for i.

NCC-IBM USER'S GUIDE       APPENDIX D                    JAN  1988


*AJ            Release a job from execution hold

*B             Backspace printer

$CJ            Cancel job in execution or awaiting execution

$DA            Display information about active jobs

*DF            Display information about jobs queued for forms

$DJ            Display job status

*DMRO          Send message to the operator

$DN            Display information about jobs awaiting execution
               or awaiting output

$Dfi            Display information about job queue

$DU            Display status information about a JES2-controlled

*E             Re-queue current activity on device

*F             Forward space device

$HJ            Place job in hold status

$1             Interrupt current activity on device

*N             Repeat current activity on device

$0             Release job from output hold

$P             Drain device

*PJ            Purge job after current activity

*R             Change job output routing

$S             Start a drained device

*T             Alter the characteristics of a JES2 controlled

NCC-IBM USER'S GUIDE       APPENDIX D                     JAN 1988


     o   The $R command will only affect  jobs  awaiting

     o   If you need to leave or close the  remote  site  and  a
         large job has already started to print, then you
         might try the following =

                $P the printer
                *I the printer


                $P the printer
                $E the printer
                  (next day)
                $S the printer
                *F printer, pages to skip

     o   On most DATA-100's with console support,  you can
         enter the commands by using a space where  the  "$"
         is shown in the commands.


JAN 1988
The following commands will list or count various groupings of
the jobs in the job queue.  Note that only those jobs with output
destined for a particular RJE station will be listed or counted
as a result o.f a command entered at that station.

Display Information About Active Batch Jobs

      Command:   $DA

      Function:  Display job numbers of all executing batch
                 jobs .

Display Information About Jobs Awaiting Execution or Awaiting

      Command:   $DN,operand

      Function:  Display job number, name, status, class and
                 priority of all jobs awaiting execution or
                 awaiting output and the percentage of spool
                 disk utilization.

      Operand:   fi=XE2,IBM

      Note:  With JES2 2.1.S and the Mellon Bank mods, $DN
             and $Dfi command operand formats changed.  Unless
             the commands are followed by ",IBM", the syntax
             will fail.

Display Information About Job fiueue

      Command:   *D2,operand

      Function:  Display the number of jobs in a particular
                 queue and the percentage of spool disk

      Operand:   e=XE2,IBM

NCC-IBM USER'S GUIDE       APPENDIX D                    JAN  1988

      Note:  With JES2 2.1.5 and the Mellon Bank mods, $DN
             and $Dfi command operand formats changed.  Unless
             the commands are followed by ",IBM", the syntax
             will fail.

Display Number of Jobs fiueued for Forms

      Command:   $DF

      Function:  Display output job information. To display
                 output for a specific job number, use
                 where xxxx is the job number.


                                        JAN 1988
The following commands serve to display and alter the status of
one or more jobs.  Only those jobs which have output directed to
the RJE station at which the command is entered will be affected
by any of these commands.
Display Status
      Function =
      Examples ••
*DJ joblist
$D 'jobname'

Display identifying and status information for
the specified job(s).



$D 'XYZ011
Place Job in Execution Hold Status

      Command:    $HJ joblist
                 $H 'jobname'

      Function:  Place one or more jobs in execution hold
                 status.   That is, those jobs will not be
                 selected for execution, printing, or punching
                 until they are subsequently released from hold
                 status.   If there are two or more jobs with
                 the same 'jobname1 specified in the second
                 form, an error message will be displayed and
                 those jobs will not be held.

                 NOTE:  This command does not affect job(s) in
                 output hold.   See the $R command.

      Examples:  $HJ5-8

                 *H 'XYZ051

Release Job from Execution Hold Status

      Command:    $AJ joblist
                 *A ' jobname'

      Function:  Release job(s) previously placed in execution
                 hold status.   If there are two or more jobs
                 with the 'jobname' specified in the second

                     APPENDIX D
JAN 1988
                 form, none of them will be held and an error
                 message will be displayed .

                 NOTE:  This command does not affect job(s) in
                 output hold .   See the $R command .


                 $A 'XYZ051
Cancel Job
                 $CJ joblist
                 *C ' jobname '

                 Cancel one or more jobs.  Each job cancelled
                 which is executing will have that activity
                 terminated immediately.  If a job is being
                 read  in or is still awaiting execution,
                 however, its  JCL will be queued for printing
                 If the job is in the output phase, it will
                 be purged. If there are two or more jobs
                 with  the 'jobname1 specified in the second
                 form,  none of them will be cancelled and
                 an error message will be displayed.


                 $C 'XYZOS'
Purge Job After Current Activity

      Command:   *PJ joblist
                 $P ' jobname '

      Function:  Purge one or more jobs after any current
                 activity is completed.  For jobs awaiting
                 execution, this command will purge all traces
                 of the job (i. e., the JCL will not be queued
                 for printing).  For active jobs, the current
                 activity (i.  e., execution, printing, or
                 punching) will be completed and the job will
                 then be cancelled. If there are two or more
                 jobs with the 'jobname' specified in the second
                 form, none of them will be affected and an error
                 message will be displayed.

      Examples :  $PJ5

                 $P 'XYZ211


JAN 1988
The following commands allow control of unit record devices
(including printers and punches) at the RJE station.  Some of
them have significant effects on the job ueing operated on by the
device specified at the time of command entry.  Some commands are
valid only for certain devices.  Both of these situations will be
noted where applicable.

Backspace Printer or Punch

      Command:   *B device   [,x]

      Function:  The command only applies to active devices.
                 It causes the data set currently being printed/
                 punched to be backspaced for the specified
                 number of pages/cards, or to the beginning
                 of the data set.  The x should be replaced with
                 the number of pages/cards to be backed up,
                 or with the character "D" which indicates the
                 data set is to be backed up to the beginning.
                 If x is omitted, one page/card is assumed.
                 If x specifies more pages/cards than have
                 already been printed, the data set is backed up
                 to the beginning.

      Examples:  *B R123.PR1,50

                 *B R100.PU1,D

Forward Space Printer or Punch

      Command:   $F device   [,x]

      Function:  This command only applies to active devices.
                 It causes the data set currently being printed/
                 punched to be skipped forward for the specified
                 number of pages or cards, or to the end of the
                 data set.  The x should be replaced with the
                 number of pages/cards to be skipped, or with
                 character "D" which indicates-- skip to the end
                 of the data set.  Printing/Punching will resume
                 with the next data set.

                 CAUTION:  If the number of pages/cards specified
                 is greater than the number of pages/cards
                 remaining in the data set to be printed/punched,
                 printing/punching  will resume with the
                 next data set if one is present (same as if
                 "D" had been specified) and the currently

                           APPENDIX D
JAN 1988
                 active data set will be lost.  If x is
                 omitted, one page/card is assumed.

      Examples:  *F R123.PR1,SO

                 $F R100.PU1,D

Interrupt Current Activity on Printer

      Command:   $i device

      Function:  Interrupt the printing/punching of the data
                 set currently being printed/punched on
                 the specified device.  The data set being
                 printed/punched is "checkpointed" and
                 re-queued.   When it is again selected for
                 output processing, the output will resume
                 at the top  of the page preceding the last
                 one printed/punched.  Ii the data set being
                 printed/punched is still the highest priority
                 data set after the interrupt command, the
                 restarting  will begin immediately.

Cancel Device


                 $C device

                 Cancel the activity on the specified devices
                 immediately.   The job being processed is not
                 re-queued.  If the device is a reader, the
                 cards remaining in the hopper are read and
                 examined.  If a JOB card is detected, reading
                 will resume with that job.  If the device is a
                 printer or punch, the next job for the device
                 will be selected, provided that there is one.

      Examples:   $c R123.PR1

                 $C R23.PU1

Re-Bueue Current Activity on Device

      Command:    $E device

      Function:   The job on the specified currently active
                 device is to  be terminated and re-queued to
                 that device.   It will not be purged as with
                 the cancel (*C) command.  If the re-queued job
                 still has the highest priority for the

NCC-IBM USER'S GUIDE       APPENDIX D                    JAN 1988

                 specified device, it will restart immediately.

      Examples:  $E R123.PR1

                 $E R23.PU1

Repeat Current Activity on Device

      Command =    *N device

      Function:  Re-queue (repeat) the job currently active on
                 the printer or punch device specified for
                 another copy to be printed or punched at a
                 later time.  The job currently being processed
                 will not be interrupted.
Drain Device
$N R123.PR1

$N R23.PU1

$P device

Prevent the specified device from being
selected for any more jobs after the current
activity is completed.  If the device is
inactive, its non-selectability will be
established immediately (it will be
"drained").  If active, it will be placed in
the "draining" status until the job being
processed is completed, ~nd will then become
"drained".  A drained device will again select
jobs once it has been started with the start
($S) command.
      Example:   $P R172.PR1

Start or Restart Device

      Command:   $S device

      Function:   Start the device specified. This command is
                 used to start devices when an RJE station is
                 first connected; to restart a device after a
                 drain or halt command; and to acknowledge that
                 a request for a forms mount is completed.

      Example:   *S R172.PR1
                              D-1 1


                                                         JAN  1988
Send a Message to the Computer Operator

      Command:   $DMRO,'message text1
                 Send the 'message text1 to the central site
                 computer operator.  If the quotation marks are
                 omitted the message will be transmitted, but
                 all blanks will be removed before it is
                 displayed.  Within the 'message text1 a single
                 quote mark is specified by using two single
                 quote marks.


Change Job Routing

      Command =
'Please re-queue job 1234 to R1231

'I canf't seem to make it work'
      Function:  Change the routing of all or part of a job's
                 output to the specified routing.  It should be
                 emphasized that this command will only affect
                 an individual job's routing.  Other jobs
                 routed to the remote will not be rerouted.
                 Once the routing is changed, the affected
                 job(s) will no longer "belong" to the original
                 RJE station, so that an incorrect
                 specification will have to be corrected by the
                 central computer operator.

                 PRT, PUN or ALL are used to specify the
                 types of output to be re-routed and represent
                 print, punch or print and punch, respectively.

                 jobid is either the complete job name in quotes
                 (e.g.  J='IIIZZ') or is the JES2 job number,
                 jjj, preceded by the levLar J (e.g. J=Jjjj)

                 nnn is the remote number, entered without
                 leading zeros.  In order to process jobs in
                 print hold, RJE operators should reroute output
                 to their own RJE site for output.  "D=LOCAL"
                 will reroute a specified job to the NCC central

NCC-IBM USER'S GUIDE       APPENDIX D                    JAN 1988

                 computer site for output processing.

      Examples:  $RALL,J=J1117,D=R121

NCC-IBM USER'S GUIDE       APPENDIX  D                     JAN 1988


     T (universal-char-set)

Note:  Operator mode used  in previous  releases  of JES2 at the
       NCC corresponds to  all parameters  left  of the slash (/)
       with the exception  of P.   An  example  follows:


NCC-IBM USER'S GUIDE       APPENDIX E                    JAN  1988


                      TYMNET ACCESS NUMBERS

Users should call NCC Network Control Facility at one of the
telephone numbers listed below to obtain the TYMNET Access
Number.  These access numbers are constantly being changed.   A
list printed in this document would always be outdated.

Note that users in the Research Triangle Park area are to use
non-TYMNET facilities (541-4642).

To report communication problems, call NCC Network Control
Facility at one of the numbers below:

                          (919) 541-4506
                          (FTS) 629-4506
                          (800) 334-0741
Compile the following information before calling to report a

     o   The telephone access number in use (or to be used)
         and the city and state locations of that number.

     o   The type of terminal and modem or coupler in use.

     o   The type of service in use (e.g., 300 BPS, 1200  BPS
         Vadic, 1200 BPS 212, 2400, etc.)

     o   The number of the port in the connection (see
         "Random 'Garbage1 Characters," below).

     o   A description of the problem.
Solving Network Access Problems

Most problems relate to access.  You can correct many of them on
your own.

NCC-IBM USER'S GUIDE       APPENDIX E                    JAN  1988


A fast busy signal indicates a temporary overload in local
telephone circuits.  Please try your call again.  If this
condition persists, notify your local telephone company.


A 'slow1 busy signal is a normal telephone-station busy signal.
It indicates that the rotary group you are dialing has reached
capacity.  Hang up and try the number again in a few minutes.  If
this condition persists, please report the problem.


Check to see that you dialed the number correctly and then place
the call again.  Please notify the Netwo^.:: Control Facility if
you have this problem while using a correct number.


Before reporting this problem, check your terminal and coupler
switch settings, cable connections, and power-on settings and be
sure the terminal is in the on-line mode.

If the same equipment has been successfully used before in con-
junction with the same telephone access number, and if you can
dial other areas and log in, please report the problem.


There are many possible sources of "garbage" characters	for
example, the use of a telephone line that goes through a switch-
board, a noise source close to an acoustic coupler, or problems
in a modem, a terminal, telephone company hardware, or TYMNET
hardware.  See "Telephone Equipment Problems" for trouble-
shooting procedures related to your own equipment.

If you can use more than one local access telephone number to
connect to TYMNET, try to eliminate the garbage characters by
disconnecting from the network and then dialing a different
number from the one you used during your preceding log in.

The most important piece of information you can supply when you
report garbage characters is the number of the port you connect
with when you first logged in.  That number is the second of the
two numbers that appear in the line of hyphens and numbers that
TYMNET displays immediately after you enter your terminal

NCC-IBM USER'S GUIDE       APPENDIX E                    JAN  1988



Drops have the same kinds of causes and solutions as garbage
characters.   Please report drops as soon as possible after they
occur if you cannot correct them.  Be prepared to report whether
you received a message when you were dropped and, if so, what
that message was.

Telephone Equipment Problems

The telephone and related equipment you use with TYMNET can cause
errors, disconnects, and other problems.  The following
precautions  are recommended:

     o   Use a telephone with a direct outside line, rather
         than going through a switchboard.

     o   Be  sure the telephone you use does not have an

     o   If  the quality of telephone service in your area is
         poor, you may need to use a data set rather than an
         acoustic coupler.

     o   If  you are using an acoustic coupler, keep it well
         away from the terminal, if possible.  If the
         coupler has a lid, keep it clos*..*..

     o   Hang up the telephone when your terminal is not in
         use.  (This can prevent the telephone's speaker and
         receiver elements from crystalizing and decreasing
         signal strength.)

     o   Be  sure that there is a piece of cotton baffling
         inside the earpiece behind the receiver in your
         telephone handset.  (Most telephone companies put
         cotton baffling in the earpiece before installing a

     o   Be  sure the terminal and modem have a circuit of
         their own.

     o   If  error rates are a problem, try a different
         telephone handset.  (Transmitter and receiver
         elements in one handset may perform better than
         those in others.)

NCC-IBM USER'S GUIDE       APPENDIX E                    JAN  1988

     o   Insure that your phone line does not have the "CALL
         WAITING" feature installed.
For explanation of a number of communication messages, see

                     TYMNET LOGON PROCEDURES

1.  From your modem/PC dial your local TYMNET access number.
    You will receive the message "ONLINE" followed by a line of
    garbled characters.

2.  At this point, you must type the character "a" and the
    network will respond with the message "please log in="
    in mark parity.

3.  Now you will need to enter the appropriate control characters
    to allocate proper buffering; the control characters do not
    echo .

M. After the control characters are entered, you must
   enter your login string followed by a carriage return.
   A login string may be comprised of a user name only, a
   user name and a subaddress, or a user name and a password.
   The following are valid login strings:

         EPA1 (1200 bps  access to IBM)
         EPA2 (2400 bps  access to IBM)
         EPACMT (TYMNET'S protocol converter)
         EPAIBMPSI (access to the IBM packet switching interface)
         EPAPSPI'-xxxx . yyyy (X.25 access to a specific host; host
                            names will be furnished upon request.)
         EPAEMAIL:.63 (access to E-Mail)

     Your  user name determines the destination point on the
     TYMNET network to which you will be routed.  If your user
     name  is entered correctly and confirmed to be valid, you
     will  receive a greeting message from the network followed
     by a  host banner and menu, at which point you will proceed
     as normal.

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN 1988



Micrographics is the process of reducing the size of an optical
image and storing the reduced image on photographic film.  The
process can be applied to printed text, to graphics, or to both.

Typically, micrographics input comes from one of two places;
source documents or computer output.  Source document
microfilming is the process of photographing a document with a
reducing lens.  In Computer Output Microfilm (COM), computer-
generated images are transferred to microfilm by a COM recorder,
which may be driven directly by the central processor in the
computer (on-line).  Usually, however, computer-generated
magnetic tape is used as output to the COM recorder in an off-
line environment.  The latter is the procedure at NCC-IBM.

The medium on which the micrographic image is preserved, usually
roll film or microfiche, commonly is called a microform.  Roll
film may be on a spool or in a cassette or cartridge, and is much
like movie film with one very important exception:  the images
may appear on the film in either of two formats, comic or cine.
In the comic format the frames are oriented parallel with the
film edge in the manner of a newspaper comic strip.  In the cine
format, the images run perpendicular to the film edge as on movie

The number of frames on a fiche and the size of the individual
frames will vary according to the format and the reduction factor
used in processing.  On the widely used 4x6-inch microfiche with
common 48X reduction factor, 270 frames can be contained in 18
columns of 15 frames each.  (See chart below.)

To aid in manual handling and identification of microfiche
contents, an eye-readable title may be printed along the top
margin of each fiche.  The title may be generated independently
or may be extracted wholly or in part from data on the fiche.
This process is discussed in more detail later.

The page header feature enables one to specify heading lines
which will appear at the top of each fiche frame.  This is
especially useful where computer-generated reports do not repeat
an identifying header constant on each report page.

Another useful feature available with microfiche is the indexing
capability.  During processing, specific data fields can be
extracted from each frame on a fiche.  This data, along with the
row and column coordinates of the frame from which it was
extracted, is recorded in the last frame, or last two frames if

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN  1988

necessary for space considerations, at the bottom right-hand
corner of the fiche.   Referring to the index frames, the user can
locate the topic he/she wants and can quickly determine which
frame on the fiche contains the desired information.

The data recorded on each microfiche can be predetermined to some
degree by taking advantage of the "fiche break" feature of the
COM equipment.  A specific data field on each frame may be
examined during processing, and a change in that field is the
signal to begin a new fiche.  The frame containing the changed
control field can be the last frame on the current fiche, in
which case the next frame will start a new fiche, or the current
frame itself can start the new fiche.  This is left to the
discretion of the user.
                  COM Procedures at the NCC-IBM

I.   Introduction

The NCC-IBM provides users with the following services:

     o   Computer Output Microforms

     o   Source Document Microforms on 16mm Roll Film

     o   Microfiche and Roll Film Duplication

II.   Magnetic Tape Handling for COM Applications

The vast majority of micrographics requests received by the NCC-
IBM are for production of computer output on microfiche.
Magnetic tape containing fixed-block, print-image report output
is the input medium for this process .

The magnetic tape prepared at the NCC-IBM for a COM application
must comply with certain prerequisites.  Standard labeled tapes
must be used, and the file to be processed must be the first file
on the tape.  The data set name for the COM file must conform to
the COM data set format:  COM.userid.acct.aaaaaaaa, where
aaaaaaaa is a 1- to 8-character name which conforms to standards
and is the same as the application name entered on the Request
for Micrographics form.

Before the tape is accepted for COM processing, it must be
catalogued. Normally, this will be accomplished by including the
following sample JCL when creating the COM tape:

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN  1988

      //OUT   DD   DSN = COM.userid.acct.name,DISP=(NEW,CATLG) ,
      //           LABEL=(1,SL,RETPD=nn),UNIT=TAPE9,
      //           DCB=(RECFM=FBA,LRECL=133,BLKSIZE=4123,DEN=3)

To catalog an existing uncatalogued data set on tape, the
following alternative is available.

      //OUT   DD   DSN=COM.userid.acct.nama,DISP=(OLD,CATLG),
      //           VOL=SER=(vol1,vo!2,. .  .),UNIT=TAPE9

To verify that the data set was successfully catalogued with  the
IEFBR14 procedure, the user should specify  the JOB CARD parameter
MSGLEVEL=(1,1) to receive the system message:

         IEF2851 COM.userid.acct.name          CATALOGUED.

If the same tape must be used as input to more than one COM
application,  multiple entries for the tape  must be made in  the
system catalog, one for each of the required applications.  The
simplest way to do this is to use IEFBR14 as shown below.
Suppose that the data on tape volume xxxxxx must  be used as input
to the three COM applications PHOTO 1, PHOT02, and PHOTOS; then
the step required to do the cataloging would be =

//DD1     DD     DSN=COM.userid.acct.PHOT01,DISP=(OLD,CATLG),
//        VOL=SER=xxxxxx,UNIT=TAPE9
//DD2     DD     DSN=COM.userid.acct.PHOT02,DISP=(OLD,CATLG),
//        VOL=SER=xxxxxx,UNIT=TAPE9
//DD3     DD     DSN=COM.userid.acct.PHOTOS,DISP=(OLD,CATLG),
//        VOL=SER=xxxxxx,UNIT=TAPE9

After the test or production application run has  been processed
by the micrographics center,  the data set will be uncatalogued
and the tape returned to the tape library.  However,  the tape
still belongs to the user until the expiration date specified in
the creation JCL.   The original tape volume serial number may be
reused for the same (or other) COM application after  COM

Twice daily,  at approximately 0900 and 1400, the  COM  application
tapes are retrieved from the tape library for micrographics
processing.  Normally, 24-hour to 48-hour turnaround  time will be
provided for production runs.  For test runs, however, the
initial setup, processing, and documentation requirements usually
require seven-workday turnaround.  The processed  microforms will
be sent to the originating user by mail, or by courier for  local
Washington, DC, users.

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN  1988

III.   Requesting Computer Output Microforms

When a user determines that his application warrants the use  of
COM, the application will enter a testing phase, whereby trial
runs are made to obtain the optimum results.  After the user  has
accepted the "best" microforms output, the COM application will
enter the production phase.  The following two sections explain
the details of these two phases.

Test Phase

Before a new or revised COM application is processed in
production mode, an initial test phase must be completed.  This
testing will ensure that;

     o   The user's COM application is accurately defined
         and documented.

     o   The user's application definition has been
         accurately transmitted to the micrographics center.

     o   The user is receiving satisfactory test COM output.

The steps to progress through the test phase follow-'

1.  The user determines that he has a COM application.  (NCC-IBM
    User Support (see Appendix A for telephone number) may be
    contacted to review microforms applicability.)

2.  The user completes a Request for Micrographics application
    form (included below).  (If assistance is required, contact
    NCC-IBM User Support).

3.  The user initiates test phase:

     a.  Sends initial application form to User Support,
         together with sample printout for application.
         (Include at least the first few (3) pages and
         examples of the fiche break field, if used.)

     b.  Tape with print-image created and catalogued for
         application (as previously described in Section
         II) .

M.  After review, User Support will contact the originating user
    and transmit the application materials to the micrographics

5.  The contracted micrographics center processes the

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN 1988

     a.  A limited number of test microforms are produced.
         It is anticipated that seven-day turnaround time
         will normally be required for processing test
         micrographics.  If the test microforms are not
         received in a reasonable period of time, the user
         should contact User Support.

     b.  If the micrographics personnel discover any
         problems during the processing of the application,
         they will contact User Support.  If necessary, User
         Support will then contact the user to help resolve
         the test problem.

     c.  The test microforms and associated Sign-Off Sheet
         are sent to the originating user.

6.  User reviews test micrographics:

     a.  If the test is satisfactory, the user returns the
         completed Sign-Off Sheet (included below) to User
         Support.  The application then enters the
         Production Phase.  (See next section.)

     b.  If the test is not satisfactory, the user should
         determine whether there is an error in the
         application form and/or the COM application
         program.  If the user cannot make this
         determination, he should contact User Support for

     c.  If the application form is in error, the corrected
         form should be sent to User Support for review, as
         described in the preceding item 3.  No additional
         sample printout will be required, but the user must
         recatalog the print-image tape.

     d.  If the application program needs modification, the
         user revises the program and creates a new print-
         image test tape and sample printout.  Once
         catalogued,  the application tape is placed in
         micrographics job queue.  A limited number of test
         fiche will be generated as described in the
         preceding item M.

Production Phase

After User Support receives the Sign-Off Sheet, the application
will be placed in production mode.   The procedures for submitting
requests for production COM applications and the receipt of COM
output follow:

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN  1988

     1.  The user has a production COM application; that is,
         the application has been through the test phase
         described in the previous section, and Request for
         Micrographics and Sign-Off Sheet forms are on file
         with User Support.

     2.  The user executes the application program that
         generates the print-image tape for the COM

     3.  The print-image tape is catalogued by =

         o  Specifying data set disposition of (NEW,CATLG) in
            the JCL when creating the print-image tape, or

         o  The IEFBR14 procedure is executed after the COM
            tape is generated (as described in Section II).

     4.  The complete COM application is processed.  A
         turnaround time of 24 to 48 hours will normally be
         required for processing the microforms.   The COM
         output is sent to the originating user by mail or
         courier run.   If the microforms are not received
         within a reasonable period, the user should contact
         User Support.
IV.   Requesting Source Document Microforms

Requests for source microfilming should be initiated using the
Request for Micrographics form, and will be subject to the same
test and sign-off procedures as were described previously for COM
applications.  On the form, check "Source Document Microfilming"
under "Service Requested" on page one, and complete the source
document microfilming section on page four.  Forward the applica-
tion, along with a sample of the source documents to be filmed,
to User Support for review and testing.  Source documents must be
"camera ready" when submitted for this process.  "Camera ready"
means =

     o   Staples removed

     o   Bindings removed

     o   Pamphlets cut into pages

     o   Pages numbered in the order in which they should
         appear on the film

     o   Inserts a title on documents which will appear on

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN 1988

After processing, the documents (along with the microfilm and the
COM Sign-Off Sheet) will be returned to the user.  When
established as production runs, source document microfilming
requests may be forwarded to HIC or NCC;  whichever is

Completed microfilm and original documents will be returned to
the user by the same mode as they were received.
V.   Requesting Microforms Duplication

The process to obtain duplication of existing microforms is much
the same as that for source document microfilming.  Check the
appropriate line on page one of the Request for Micrographics
under "Service Requested" and forward it with the original fiche
or film to be copied.  Only one copy will be made unless
otherwise indicated by the user in the "Output" section of the
request form.

Special Instructions on Filling Out Request for Micrographics
Form ••


A title, readable with the naked eye, can be printed along the
top margin of each fiche.  The wording can be supplied by the
user,  or specified fields can be extracted from the data and
inserted into the title, or a combination of both sources can be
employed.  From the "Title Extract Fields Data1 table it can be
seen that extract fields are specified by line number, field
length, and position within a line.  The line specification must
account for the number of lines used for page headers as covered
in the following section "Indexing".  Allowing for page header
lines is necessary since the frame print line count starts from
the first page header, if any.  "Beginning Print Position" in
this case is counted from the first printable character—the
carriage-control character is not included.   The maximum
permissible character count for title lines is determined by the
reduction factor employed, and can be found in the following

         24X—80 characters on 2 lines
              40 characters on 1 lines

         42X--128 characters on 2 lines
               64 characters on 1 line

         48X--144 characters on 2 lines
               72 characters on 1 line

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN  1988

Normally, field extractions are done from the first frame each
time a new fiche is begun.  The user has the option to "hold" the
extractions, in which case the same extract fields will appear on
every fiche and will come from the first data frame of the first
fiche only.  This is useful if data (such as a date) appears  only
on the first frame of the job.


To facilitate finding information on the fiche at a later date,
information may be supplied to build the fiche index.  Fields of
data may be selected which will identify the contents of each
frame.   The index fields and the row and column coordinates of
the frame from which they were taken will be used to build the
index in the last frame of the fiche.  Up to four separate
fields, or subkeys, may be specified for each index entry.  For
each subkey, the position within the print line, disregarding the
carriage-control character, and the subkey field length must  be
supplied.  The sum of the lengths of all subkeys used may not
exceed 60.  A particular subkey may appear on different lines in
different frames.  If this is the case, a range of lines to be
searched can be indicated.  The mask data tells what the
character makeup of the subkey is, as can be seen in the mask
character key beneath the index table.  Note that the starting
print position and field length of the mask may be different  than
those entered for the subkey.  This enables the user to specify a
unique part of the field as the mask, and to extract the entire
field for use in the index.  For example, a given 7-character
field might vary in its alphanumeric configuration, but the
fourth character is always a hyphen.  The entire seven characters
would make up the key field, but the mask upon which the search
is made would be the one-character hyphen in the fourth position
of the key.  A mask character of "S" specifies any non-blank
character other than alphanumeric.

Fiche Break

Controlling the point at which a new fiche is begun is a
capability provided by the fiche-break feature of CON processing.
The fiche break logic is similar to, but more complex than, the
index logic.  With the fiche break, specific conditions are
established as the criteria for completing one microfiche and
beginning another.  In this manner it is possible to predetermine
what will be on each fiche, rather than allowing the equipment to
make the break simply because a fiche has been filled to
capacity.  The fields specified under "Fiche Break Requirements"
on the NCC-IBH Request for Micrographics may be examined for
either of two conditions:  either the field changes, or it
matches pre-established criteria.

A maximum of four break conditions may be specified on any one

NCC-IBM USER'S GUIDE       APPENDIX F                    JAN  1988

application.  A maximum of two match conditions, or four change
conditions, or a combination of both, up to four total
conditions, may be used.  The occurrence of any one of the
indicated conditions will cause a fiche b^eak.  The line
containing the break field may be in either the last frame on the
current fiche or the first frame on the new fiche, at the option
of the user.

Change breaks are limited to fixed fields of not more than six
characters each.  A fixed field is one that always occurs at the
same print position on the same line number.  Please remember
that the carriage-control position is ignored when determining
"Beginning Print Position".

Match break fields may be fixed or may "float" within a range of
line numbers.  The beginning print position may not vary.  For
each break condition, up to six match fields may be specified.
When the designated break field matches any one of the six match
field alternatives, a fiche break will be executed.  The maximum
length for a match field is 36 characters, but the maximum value
for the sum of the lengths of all match fields is also 36.
Therefore, if a field 36 characters long were selected, the
application would be limited to that one field.  However, six
fields of six characters each could be used as another

I                                   I  RETURN COMPLETED REQUEST  TOM
I MICROGRAPHICS CENTER  USE  ONLY     I  Washington Information Center
IDATE RECEIVED 	I  401 M Street, SW             |
ICONTROL NO.:  	|  Attention:  Micrographics    I
I                                   I  Washington, DC  20460        I
                         USER  INFORMATION
                                           ZIP CODE _
                  PHONE NO. 	 ROOM NO. 	
                  USER I.D. 	 ACCOUNT NO
                        TYPE  OF  APPLICATION

                     .NEW  APPLICATION (TEST STATUS)
                     CHANGE  APPLICATION
               CURRENT  STATUS: 	 TEST

                         SERVICE  REQUESTED
                     .ORIGINAL  MICROFICHE
                     .ORIGINAL  ROLL  FILM
                     .DUPLICATION  OF MICROFICHE
                     DUPLICATION  OF ROLL FILM
                        SPECIAL  INSTRUCTIONS
MICROGRAPHICS                                          Page  1 of 5


 INPUT (Required for microfiche or roll film request)


    	TRACK (7 or 9)
    	BPI (556,800,1600 ONLY)
    	CARRIAGE CONTROL POSITION (if exist, usually pos.  1)
          carriage control exist, usually pos. 2)
 VERTICAL TABBING =  Used only for special formatting of page
                   frames.  ("Channel" number is found in
                   carriage control position; when channel
                   number is detected, related lines will
                   print on line number specified.)

I 1 I 2 I 3 I 4 I 5 I
1 1 1 1 1 I
i i I I I I
1 1 1 1 1 1
i i i i i i i i
1 1 1 1 1 1 1 1
i i ; i i i i i
MICROGRAPHICS                                         Page 2 of 5


 OUTPUT(continued from page 2)

I                                                               !

I                                                               I
I                                                               I
     	_______________	___	_	 J

                        TITLE  EXTRACT  FIELDS DATA
                    FIELD     123456
                           p—————- —_________.«» ^m	—      -|





Skip lines
after linel
Skip lines
after Iine2
Skip lines
after lineS
(Skip lines
after Iine4

                              Page  3  of 5


 OUTPUTC continued from page 3)




' I








  MASK CHARACTERS:  Numeric = 9; Alphabetic = A; Alphanumeric  =X,
                    Special Character = S; Literal = L.

       FORMAT              REDUCTION RATIO
       	 COMIC             	  24X
       	 CINE              	  25X
                  CARTRIDGE TYPE
                      Page 4 of 5



                         COND 1    COND 2   COND 3












  Page 5 of 5

CONTROL NO.  	                        DATE
                                                ZIP CODE
           PHONE NO. 	  ROOM NO.

                        SIGNATURE INDICATES APPROVAL
        *If test results are unsatisfactory, resubmit a NCC-IBM
        Request for Micrographics form noting changes desired.
        Do not return this form.  For assistance contact WIC
        at 202 488-5900.  Return this sheet to WIC only when
        application is approved to be placed in a production


                 EPA Washington Information Center
                 401 M Street, SW
                 Washington, DC  20024
                 Attention:  Micrographics
                            Page F-15

NCC-IBM USER'S GUIDE       APPENDIX G                    JAN  1988


                  Programming Language Software

               Language                    Version

                Assembler (H)                 2.1

                BASIC (VS)                    1.1

                COBOL (VS)                    2.0

                COBOL (OS/VS)                 1.2.4

                FORTRAN VS                    2.2.0

                PL/1 (Optimizer)              1.5.0

                      Applications Software

               Package                     Release

                ADABAS                        4.1.9

                BASIS                           J

                BMDP                          1987

                CALCOMP BASIC SUB             71

                C-TAB II                      2.3

                CUECHART                      2.0

                DB ACCESS II                  2.0

                EASYTRIEVE+                   10

                ESP                           1980

                FOCUS                         5.0.4

                GEOMAP                        3.0

                GLIM                          3.0

                IMSL                          10.0

                            Page G-1

NCC-IBM USER'S GUIDE       APPENDIX  G                     JAN 1983

                INTERORG                       5.1

                IRS                            2.8

                LIBRARIAN                      3.6.0

                MARK IV                        8.0

                MEGACALC                       14.0.2

                MPS III MATH.  PROG.  SYS.       6317

                NATURAL                        1.2

                OPTIMIZER III                  14.2

                PANVALET                      11.0

                SAS                            5.16

                SAS/GRAPH                      5

                SCRIPT                         3.7

                SURFACE II                    1978

                SYNCSORT                       2.5

                SYSTEM 2000                   11.5

                Tektronix IGL                  6.2

                Tektronix TCS                  3.0

                Tektronix AGII                 1.2

                Tektronix Preview              1.0

                TELL-A-GRAF                    6.0

                TELLAPLAN EXPERT               1.2
                            Page  G-2

NCC-IBM USER'S GUIDE       APPENDIX H                    JAN  1988


For Printing in upper case only to SYSOUT=A=


          /*JOBPARM LINES=111
          //   EXEC PRINTSE2,LIB='data-set-name1,DEST = RMTnnn[ ,COPIES


          PRINTOFF 'data-set-name1  bbbb DEST(RMTnnn) [COPIES(m)]

For Printing on IBM 3800 at NCC in upper and lower case on
3-hole notebook paper:

     BATCH:   (this also specifies that output be burst and trimmed)

          //. . .JOB. . .
          /*JOBPARM LINES=111
          //stepname EXEC PRNTEXT,LIB='data-set-name'[,COPIES=m]

     ISO :

          PRINTOFF  'dsn1 bbbb TEXT FORMSC6371) CLASS(F) [COPIES(m)]


data-set-name is the fully qualified name for a sequential
data set (e.g.,  JUSD.IBM.DOCUMENT.DATA or the name and member
name for a partitioned data set (e.g., JUSD.USERS.GUIDE(MANUAL)).

 Ill is estimated lines printed in thousands; default is 20
     (i.e.,  20, 000 lines).
 nnn is remote ID.
   m is number of copies desired.
bbbb is bin number for distribution.
                            Page H-1


                    APPENDIX I-'  ROUTE CODES
                                                 JAN 1988
Note:  The following codes are used in the Route Code column to
indicate destination:


     To be distributed to Washington, DC area user by
     courier at the user's expense.

     Hold at WIC for user retrieval.

     To be distributed via Express Nail, Priority 1,
     at the user's expense from RIP.

     To be distributed via U.S. Mail, Fourth Class
     from RTF.
All jobs to be mailed (route code M) or sent by Express Mail
(route code F) should be routed to NCC central site in order
to be shipped by DPSS.  This is the most expeditious means.
          National Computer Center bin at RTF for user
          retrieval or RTF area courier delivery.
          Replace nnn with bin number assigned by DPSS.

          Fairchild Building-EPA,  Second Floor West, Room 6,
          499 South Capital Street, SW, Washington, DC  20460

          Fairchild Building-EPA,  Third Floor East, 499
          South Capital Street, SW, Washington, DC  20460

          PRC, 1505 Planning Research Drive, McLean, VA
          Crystal Mall, EPA, 1921 Jefferson Davis
          Highway, Building Two, Room 1121, Arlington, VA
          U.S. Geological Survey, Water Resources
          Division, Sunrise Valley Drive, Room 6-A301F,
          Reston, VA  22091

          Computer Science Corporation, 6521 Arlington
          Boulevard, 2nd floor, Room 210, Falls Church,
          VA 22046
          Unisys, 7929 West Park Drive, Room 252, McLean,
          VA  22101
                            Page 1-1

NCC-IBM USER'S GUIDE       APPENDIX I                     JAN  1988

C       807       Science Application International  Corp.,
                  Suite 800,  1710 Goodridge Drive,
                  McLean, VA  22102

C       808       EPA, Fairchild Building, 499  South Capital  Street,
                  SW, 8th Floor, ATTN =  Joe Showman,  Washington, DC

C       809       JRB Associates, 8400 West Park  Drive,
                  First Floor Receptionist, McLean,  VA  22102

C       810       VERSAR, Incorporated, 6850 Hechinger  Drive,
                  Springfield, VA  22151

C       811       Centec Building, 11260 Roger  Bacon Drive, Suite
                  300, Reston, VA  22091

C       812       Horizon System Corporation,  1800 Michael  Papaday
                  Drive, Reston, VA  22091

C       813       Montgomery  County Exec. Bldg.,  101  Monroe Street,
                  6th Floor Receptionist, ATTN: S. Ambush,
                  Rockville,  MD

C       814       E.H. Pechan Associates, 5537  Hempstead  Way,
                  Springfield, VA  22151

C       815       AMS, Inc.,  7th Floor, 1777 N. Kent Street,
                  Arlington,  VA  22209, Attn=  J.J. Wind

C       816       VIAR and Company, 300 North  Lee Street, Suite
                  200, Alexandria, VA  22314

C       817       (BFSD) Crystal Mall-EPA, 1921 Jefferson Davis
                  Highway, Building Two, Room  700, Arlington,  VA

C       818       World Trade Center, 400 Virginia Ave.,  SW,
                  Suite C-20, Washington, DC 20024

C       819       Viar £ Company, 209 Madison  Street,  Alexandria,
                  VA 22314

C       820       Electronic  Data Systems, 6430 RocXledge Drive,
                  Bethesda, MD 20817  Attn=  Mrs. Green.

C       821       Computer Science Corporation, 8100  Gatehouse
                  Road, Falls Church, VA 22042

C       822       Fairchild Building-EPA, Third Floor  Room  5,
                  499 South Capital Street, S.W., Washington,  DC
                            Page 1-2

                   APPENDIX I
JAN 1988
          System Control Branch, 19th £ E Streets, NW,
          Room 5431, Attn=  Gardner Lee,
          Washington, DC 20M60

          Crystal City 2, 1921 Jefferson Davis Highway,
          Room 1002B, Attn=  Doretha Hillsman, Arlington,
          VA  22212
        825       Crystal Mall, 1921 Jefferson Davis Highway,
                  Room 1022, Arlington, VA  22212

        826       Compex Corporation, 5500 Cherokee Avenue,
                  Suite 500, Alexandria, VA 22312
                  Attn =  Bill Sprow

        xxx       Hold at the WIC (EPA Headquarters) for user
                  retrieval.  Replace xxx with the bin number
                  assigned by WIC.

        User-ID   To be distributed by DPSS-RTP personnel via
        Initials  Airborn Mailress, Priority  1 (overnight service),
                  at the expense of the user  represented by the
User-ID   To be distributed by DPSS or I/O Control
Initials  personnel via U.S. Fourth Class at RTP, to
          the user represented by the User-ID initials.
                            Page 1-3

NCC-IBM USER'S GUIDE       APPENDIX  J                     JAN 1988

Samples using graphic software packages  are  only  available in the
version of this manual that is printed in  December.
                            Page J-1

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN 1988

Through the facilities of the EPA National Computer Center (NCC),
the National Data Processing Division provides ADP training at a
reasonable cost to EPA, and to other Federal agencies and state
and local governments.   Classroom offerings are scheduled
regularly at the NCC in Research Triangle Park, NC, and in the
Washington Information Center at EPA Headquarters.  Most courses
can also be provided at any location specified by the requestor.
NCC Training supplements classroom instruction by making
videocassette courses available to users without direct charge,
and through computer-based training via PHOENIX on the NCC-IBH
system.  For further information about any training services,
contact =

     EPA National Computer Center
     Training Office--MD3MB
     Research Triangle Park, NC  27711
     (919) 541-7862   (FTS) 629-7862
                   ***  GENERAL INTEREST  ***


DESCRIPTION:   This course assumes no prior knowledge of computers
or programming.  It is designed to give the participant an
understanding of the fundamental concepts of automatic data
processing.  Topics include the historical development of
computers, the impact of computers on society, basic computer
terminology and concepts, identification  of hardware
components, preview of programming languages and applications,
and an introduction to EPA's National Computer Center.  An
overview of some of the major Agency ADP applications is also

LENGTH:   1 day

                            Page K-1

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  This course presents a comprehensive overview of
the adult education process as it applies to skills training.
The format is oriented toward active student participation in the
form of discussions and exercises.  The course will culminate in
lesson preparation and presentation by all participants.

LENGTH:  3 days


                    ***  NCC MAINFRAMES  ***


DESCRIPTION:  This course is designed to familiarise new users,
or those who have no previous experience with IBM systems, with
the NCC-IBM operating environment.  The emphasis is on use of
ISPF as the primary means to accomplish day-to-day data
processing work.  Some time will be spent discussing fundamental
elements of IBM processing, like data set naming conventions,
file creation and maintenance, and use of Job Control Language.

LENGTH:  2 days



DESCRIPTION:  An overview of the capabiliites and structure of
ISPF are presented in this course.  Topics include a discussion
of ISPF menus,  ISPF primary commands and on-line commands, job
retrieval capabilities, and data set management utilities.

LENGTH:  1 day

PRERE2UISITES:   INTRODUCTION TO ADP or equivalent knowledge and
some experience in a non-full-screen IBM environment.
                            Page K-2

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  For users who are inexperienced in IBM Job Control
Language, an overview is presented of the IBM OS/VMS operating
environment as installed at the NCC.   Basic features of the
operating system are explained, including the JOB, EXEC, and  DD
statements as well as the use of catalogued and private
procedures.  The JES2 control statements are discussed, and
interpretation of the JES log messages and STEP/JOB summaries are

LENGTH:  2 days

equivalent experience.


DESCRIPTION:  This course is intended for full-screen users who
need an overview of the TSO command language.  Techniques are
presented that make procesing under TSO more efficient, and the
basics of CLIST writing are explained.

LENGTH:  1 day

equivalent experience.

DESCRIPTIONS:   This course introduces the tools that are
necessary for productive data processing using the TSO command
language in a dial-up mode.  The most commonly used TSO commands
will be presented, along with the TSO line editor and the
concepts of logon procedures.   The writing of TSO command
procedures is also discussed.

LENGTH:  3 days

PRERE2UISITES:  INTRODUCTION TO ADP or equivalent experience.
                            Page K-3

NCC-IBM USER'S GUIDE       APPENDIX K                     JAN  1988


DESCRIPTION:  This course is aimed at programmers who  are
familiar with the computer systems of other manufacturers,  but
who are new to the IBM TSO/ISPF operating environment.   A  brief
overview is presented of ISPF and the TSO command language, and
Job Control Language and job tracking/retrieval via  SDSF  are  also

LENGTH:  3 days

PREREBUISITES=   Proficiency using a non-IBM computer system.


DESCRIPTION:  This course provides an in-depth look  at TSO
command procedures (CLIST's) and introduces their role in
applications.  CLIST variables, CLIST statements, advanced
topics, and CLIST applications are covered.

LENGTH:  1  day



DESCRIPTION:  TSO/ISPF users are introduced to two additional
editing capabilities:  ISPF Edit Macros and File Tailoring.   ISPF
messages are also discussed.

LENGTH:  1  day



DESCRIPTION:  This course is intended for experienced  ISPF  users
who are interested in developing full-screen, interactive
applications using the ISPF dialog manager.  Dialog  variables,
panel definition, messages, table management, and ISPF services
will use command procedures to drive ISPF dialogs.

LENGTH:  2 days

                            Page K-4

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  This course introduces the Logical Mainframe
concept, and describes the LMF system and operating environment.
The use of TSO/ISPF, SDSF, JCL, and ISPF utilities to perform
data processing tasks on the LMF are addressed, and an
introduction to CLIST writing is presented.  Extensive use of
hands-on exercises reinforces the material for participants.

LENGTH:  5 days



DESCRIPTION'-  The functions and capabilities of IBM's Resource
Access Control Facility (RACF) are presented in this course.
Participants uill learn how to use RACF to protect their data
sets residing on IBM mainframes by limiting the number of users
who may access them and by restricting the type of access a given
user or group of users may have.

LENGTH:  1 day

PRERE2UISITES=  Participants should be active users of TSO/ISPF.


DESCRIPTION:  TELL-A-GRAF's capabilities are introduced.  A
sampling of the TELL-A-GRAF commands, representing each of the
four TELL-A-GRAF "personalities" (PLOT, BAR, PIE, and TEXT), is
examined in detail.  Topics also include printing and saving
charts, alphabets, fonts, page layout, file overview, and hands-
on practice sessions.  This class is for beginning TELL-A-GRAF

LENGTH:  i day

                            Page K-5

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  This course is designed for FOCUS users who will  be
accessing pre-existing data bases and applications.  The emphasis
of the course will be on creating custom reorts and graphs  from
existing data files.  FOCUS statistics and TED, the FOCUS editor,
will be introduced, also.

LENGTH:  2 days

PREREQUISITES:  Solid experience using ISPF and TSO or completion
of the Introduction to IBM Full Screen Processing course.


DESCRIPTION:  Participants in this advanced course must have
prior experience with TSO/ISPF and FOCUS.  The course is designed
for users who will be planning and creating data bases and
applications within FOCUS.  The major topics include FOCUS  file
creation, external file processing, maintenance procedure design,
and internal execution.

LENGTH:  2 days

PREREfiUISITES :  The courses FOCUS Reporting and Introduction  to
IBM Full-Screen Processing or equivalent experience.
                       ***  LEXITRON


DESCRIPTION:  Beginning word processing on the Lexitron equipment
is presented through a series of hands-on exercises.  The course
is designed to acquaint the user with Lexitron usage, and to
explain the available tools that perform basic word processing
tasks .

LENGTH:  1 day

PRERE2UISITES:  Typing skills.
                            Page K-6

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  Advanced word processing capabilities are covered
in this module.   Practical exercises centered on menu pages and
special function keys are used to explore the full range of word
processing capabilities.

LENGTH:  i day

experience .


DESCRIPTION:  Fundamentals of Lexitron operations and an
introduction to word processing comprise the first half of the
course.  The afternoon is devoted to telecommunications
capabilities and interfacing with mainframes .

LENGTH:  1 day


DESCRIPTION:  This course provides a brief overview of the PC
hardware and software supported by EPA and an introduction to
CROSSTALK, the communications package.  Special emphasis will be
placed on the DOS operating system.

LENGTH:  1 day


DESCRIPTION:  The concept of a spreadsheet and its real world
applications will be introduced.  The course will include basic
cell manipulation, file handling, and spreadsheet exercises.

LENGTH:   1  day

                            Page K-7

NCC-IBH USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  The course will familiarize the LOTUS  1-2-3 user
with the data base and programming capabilities of the language.
All data management commands will be covered, and the writing of
macros will be explained.

LENGTH:  1  day

PRERE2UISITES:  Working knowledge of LOTUS  1-2-3 worksheet, file,
and print commands.


DESCRIPTION:  This course focuses on the fundamental purposes and
capabilities of a data base management system.  Data base
creation, editing, searching, and reporting will be explained
using practical examples.

LENGTH:  1  day



DESCRIPTION:  The dBASE III user will become familiar with the
memo field, and with the use of memory variables, command files,
and programming commands.  The building of  a menu program and
customizing of input screens will also be addressed.

LENGTH:  1  day

knowledge and experience with dBASE III.


DESCRIPTION:  The functions and capabilities of the latest
version of  the dBASE family of software packages will be
introduced.  Hands-on exercises will be used to provide maximum
reinforcement of the concepts presented in  this course.

LENGTH:  1  day

                            Page K-8

NCC-IBJ1 USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  The fundamentals of WordStar, the PC word
processing package, will be covered in this course with emphasis
on creating, editing, and saving WordStar files.

LENGTH:  1 day



DESCRIPTION:  Participants will learn the basics of micorcomputer
hardware and DOS (the Disk Operating System).  WORDPERFECT funda-
mentals will be covered including creating and editing documents,
making format changes, and saving documents to disk.  Spelling
check,  search and replace, page and line formatting are also

LENGTH: " 2 days



DESCRIPTION:  This course is an introduction to the fundamentals
of WORDPERFECT.  Participants will learn to create and edit
documents, to make specific formatting changes, and to save
documents to disk.  Spelling check, search and replace, and page
and line formatting features are also covered.

LENGTH:  i day

PREREBUISITES:   Working knowledge of the PC or INTRODUCTION TO


DESCRIPTION:  Advanced WORDPERFECT features will be covered
including macros, windows, mail merge, automatic outlining and
table of contents generation, footnoting, super/subscript,
extended character set, block protect, columnar text, default
setup menu, document summary, hyphenation, and thesaurus.

LENGTH:  1 day

                            Page K-9

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  This seminar is intended to provide an overview  of
the functions and capabilities of personal computers--uith
emphasis on the standard EPA PC conf iguration--f or
administrators, managers, and others who may have to make
decisions about the acquisition and use of PC's in their
organizations.  Participants will get a hands-on introduction  to
the hardware and software through a series of demonstrations and
exercises dealing with each supported software package.

LENGTH:  1 day



DESCRIPTION: This course provides an introduction to the EPA
PRIME computer system.  Featured are the hardware, common
software, the menus, communications with other systems, ELINK  (a
local electronic mail system similar to EMAIL), and additional
menu options.  Basics of the operating system, PRIMOS, are
covered to include the handling of files, a few PRIMOS commands,
and the Editor .

LENGTH:  1 day



DESCRIPTION: The fundamentals of TEXT, one of the word processing
packages on the Prime, are covered in this course.  The emphasis
is on creating,  editing, and saving TEXT files.  Some advanced
text processing features are also covered, including the
abbreviation function, spelling checker, and file merger.

LENGTH:  i day

                            Page K-10

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  This course will help new users become proficient
using INFO, a fourth-generation language and data base management
system.  Basic features of INFO are explained, including data
base creation, data retrieval, data management, reporting
writing, and an introduction to programming.

LENGTH:  2 days



DESCRIPTION:  For those who will be programming in INFO, this
course builds on the material presented in INTRODUCTION TO INFO.
The emphasis is on sound programming concepts and commands.  Some
time will be spent reviewing INFO commands, introducing commands
that provide more efficiency in INFO, and accessing existing

LENGTH:  3 days



DESCRIPTION:  This course introduces the PRIME'S graphics
management tool.  Over 100 predefined PIE, BAR, LINE and FREE
FORM PLOT-FORMS are included for immediate plotting.   The course
also addresses the sophisticated change editor within Versagraph,
which allows restructuring of the standard platform for
customization of graphics output.

LENGTH:  2 days

                            Page K-11

NCC-IBM USER'S GUIDE       APPENDIX K                    JAN  1988


DESCRIPTION:  This course discusses INFO-TEXT, an addition to
INFO that allows variable-length text records to be created in  a
data file by invoking the word processor WordMARC.  Topics
covered include data file creation with INFO-TEXT attributes,
invoking WordMARC to add and update records, displaying INFO-TEXT
data at the terminal, and searching INFO-TEXT records for
specific text strings.

LENGTH:  1  day



DESCRIPTION: 20/20 is a multiwindowed spreadsheet package very
similar to  LOTUS 1-2-3.  This course teaches building,
formatting, editing, printing, and the storage of worksheets.
Demos and discussions of the advanced features of 20/20 include
graphics, goal seeking, worksheet linking and consolidation,
project modeling, and data base capabilities.

LENGTH:  1  day



DESCRIPTION:  WordMARC is a menu driven, multilanguaged, full
screen word processing package that features extensive
dictionaries, both global and personal; document encryption;
scientific  word processing capabilities; and easy interface to
INFO.  This course includes word processing basics and advanced
features of spell checking, merging documents, and the
abbreviation glossary.

LENGTH:  1  day

                            Page K-12

NCC-IBM USER'S GUIDE       APPENDIX K                    OAN  1988


DESCRIPTION:  VAX Course 1 is designed to familiarize the end
user with the basics for using the VAX/VMS computer.  Operating
system commands will be covered, as well as the essentials of
constructing command procedures for repeated usage.  The VAX
batch environment, editing, and NCC communication basics are
detailed as part of the overall presentation.

LENGTH:  1 day



DESCRIPTION:  This course is designed to expand the knowledge of
the VAX/VMS user.  Concentration areas include additional
information on system usage:  Advanced Command Procedures, the
VAX Librarian Utility, Task Spawning, the Run Time Library (RTL),
System Services, and the Symbolic Debugger.  The new Extensible
VAX Editor (EVE) will be discussed as a powerful addition to the
user's environment.  VAX FORTRAN will be covered as the language
of choice for examples, as well as separate presentations on the
FORTRAN Compiler and special features of the language.

LENGTH:  1 day



DESCRIPTION:  This course parallels the equivalent course for the
IBM mainframe.  It is designed for the user needing to produce
reports from existing data bases.  The nature of data base
construction and maintenance will be covered in a subsequent
course.  Topics for this course include standard reporting,
advanced reporting, and short section on statistics and data
graphing.  Also covered are creation of new variables and the use
of batch and command mode for running FOCUS reports.

LENGTH:  2 days

PREREfiUISITES:  VAX Course 1 or equivalent experience.
                            Page K-13


JAN 1988
NCC Training provides an online training capability on the NCC-
IBM mainframe via the PHOENIX computer-based training system.
PHOENIX is available under TSO to any IBM user having an IBM 327x
or 3270 compatible terminal capable of full-screen display.

The available courses and a short description of each are listed
below.   If you are interested in any of the courses, please
contact NCC Training at (FTS) 629-7862 or (919) 541-7862 for
student ID and workbook.


A4095     Basic MVS JCL

A4096     MVS JCL for Programmers

A5547     CLIST

A5560     ISPF/PDF Dialog Manager for MVS

A5566     Fundamental ISPF/PDF

A8117     Screen Design

A8123     Basic Computing Concepts

A8125     3170/3270 Keyboard Skills

A8133     TELL-A-GRAF

A8137     Introduction to SAS

A8138     SAS= Advanced Topics

A8139     FOCUS Reporting

A8141     SAS/FSP

A8143     SAS/GRAPH

A8144     SAS/ETS

A8153     Using the Terminal

A8154     Basic COBOL Language

A8158     Intro to Easytrieve Plus
                            Page K-14

NCC-IBM USER'S GUIDE       APPENDIX L                    JAN 1988

This alphabetical list contains the NCC informational data sets
available for users.  Note that JUSD is the NCC data set name
qualifier for on-line documentation.
- A -

- B -

- C -
                            Page L-1

NCC-IBM USER'S GUIDE       APPENDIX  L                     JAN 1988

- D -

- E -

- F -

- G -

- H -
                            Page L-2

NCC-IBM USER'S GUIDE        APPENDIX L                     JAN 1988

- I -

- J -

- K -

- L -

- M -
                             Page L-3

NCC-IBM USER'S GUIDE       APPENDIX L                    JAN 1988

- N -

- 0 -

- P -

- 2 -

- R -
                            Page L-M

NCC-IBM USER'S GUIDE        APPENDIX  L                    JAN  1988

- S -

- T -

- U -

- V -
                             Page  L-5

NCC-IBM USER'S GUIDE       APPENDIX L                    JAN 1988

- W -

- Z -
                            Page L-6

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN  1988

abend;   From "abnormal end".  The abnormal termination of a
     program's execution due to a program error, an I/O error, a
     hardware failure, or the detection of an illegal condition.
     On an IBM 370, an abend will cause a core dump if an
     appropriate SYSUDUMP DD-card is included, and will inhibit
     the execution of any following job steps which are not
     specifically exempted from this action.

access:   (1)  The manner in which files or data sets are referred
     to by the computer.  (2)  In RACF, the ability to obtain the
     use of a protected resource.

access authority:   An authority that relates to a request for a
     type of access to protected resources.  The access
     authorities are NONE, READ access, CONTROL access (for VSAM
     data sets), and ALTER access.

access environment:   A description of the current user, including
     User-ID, current connect group, user attributes, and group
     authorities.  An access environment is constructed during
     user identification and verification.

access list:   A list within a profile of all authorized users and
     their access authorities.

access method:   The manner and/or the software used by a computer
     to reference files of information, such as sequential,
     random, etc.

access time:   Usually, the time to access a data item—That is,
     the time which elapses between a request for a particular
     record and the beginning of the read or write operation which
     satisfies that request.  Less frequently, denotes the time
     which elapses between a request for a particular record and
     the end of the read or write operation which satisfies the

account code:   At the NCC, a four-character code used for the
     billing of computer services and other miscellaneous charges.
     The account code identifies the allocation or suballocation
     of funds to be charged, while a user's User-ID identifies a
     specific individual.

address space:   The virtual storage assigned to a batch or
     terminal job.  Each address space consists of the same range
     of addresses.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

address space identifier:   A unique, system-assigned identifier
     for an address space.

address translation:   The process of changing the address of an
     item of data or an instruction from its virtual address to
     its real storage address.

ADP:   Automatic Data Processing, having to do with computerized

ALC:   Assembler Language Coding, the symbolic coding language
     used for machine-level programming of IBM 360 and 370 series

alphanumeric:   Consisting of characters, including the 26
     alphabetics and the 10 numerics and often the space
     character;  sometimes means any character in the computer's
     character set.

archive:   TO remove (a data set) from an on-line or active storage
     medium or volume to an off-line or inactive storage medium or

ASCII:    American National Code for Information Interchange.
     The standard code, using a coded character set consisting
     of 7-bit coded characters  (8 bits including parity check),
     used for information interchange among data processing
     systems, data communication systems, and associated equip-
     ment.   The  ASCII set consists of control characters and
     graphic characters.

ASM2:     Automated Space Management System, a general-purpose,
     direct-access, space management system.  It provides
     facilities  for archival and retrieval, backup and recovery,
     migration and volume configuration, storage maintenance,
     and ASM2 maintenance in OS and OS/VS systems.

Assembler:    A processing program that transforms symbolic
     machine-level programs (such as those written in ALC) into
     actual machine code.

asynchronous:   A mode of data  transmission between data
     processing  devices where one operation must be fully
     completed before another operation is initiated;
     e.g. a TTY  mode

authorization checking:   The action of determining if a user is
     permitted access to a RACF-protected resource.

authorized  program:   A system  program or user program that is
     allowed to  use restricted  functions.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988
automatic mode:    The setup and selection of jobs on a printer
     controlled  by JES, as opposed to being controlled by the

auxiliary storage management:    A facility that controls the
     allocation  and release of pages on external page storage, and
     schedules  I/O operations  on the page data set.

background job:    (i)  A low-priority job, usually a batched or
     noninteractive job.  (2)   Under ISO, a job entered through
     the SUBMIT  command or through SYSIN.  Contrast with
     foreground  job.

BAL:    From Basic Assembler Language.  Sometimes used to denote
     the Assembler Language for the IBM 360 and 370 computers.
     Originally, BAL implied the Assembler Language exclusive of
     its macroprocessing capabilities and ALC implied the
     inclusion  of those capabilities, but the two terms have
     become practically synonymous.

basic control (BC) mode:   A mode in which the features of a
     System 360  computing system and additional System 370
     features,  such as new machine instructions, are operational
     on a System 370 computing system.   See also extended control
     (EC) mode.

batch:    A term  applied to computer jobs that are submitted for
     processing  at some later  time;  implies that the job does not
     interact with a low-speed terminal during its execution (cf.

batch processing:   (1)  Pertaining to the technique of executing
     a set of computer programs such that each is completed before
     the next program of the set is started.  (2)  Pertaining to
     the sequential input of computer programs or data.  (3)
     Loosely, the execution of computer programs serially.  (M)
     Under ISO,  the processing of one job step in a region, so
     called because jobs are submitted in a group or batch.  (5)
     See also execution batch  processing.

baud  rate:   The transmission  rate of a data communication line
     (or channel) measured in  bits per second.

BC mode-'   Basic control mode.

BCD:    Binary Coded Decimal, a data coding scheme using six
     information bits, four "numeric," and two "zone" bits and
     usually accompanied by a  "check" bit for parity.

benchmark:   A  job or series of jobs executed on more than one

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN  1988

     computer system to compare the processing abilities of those
     systems; most frequently the speed and cost with which the
     benchmark is completed.

binary:   The base-two numbering system.  In binary, each digit
     may have only the values zero or one; accordingly, each  digit
     may be represented by a single bit.  Most modern computer
     systems represent numerical information in binary.

bit:   A binary digit, either 0 or 1.

block:    A group of logical records written and read together for
     speed and storage efficiency.

bpi:   From bits per inch.  The number of longitudinal bits,  or
     frames, per inch of magnetic tape.  Used following a number
     to express the density of data recording on magnetic tape.
     In general, the greater  the density, the greater the speed and
     reliability of data transfers.

broadcast data set:   Under TSO, a system data set containing
     messages and notices from the system operator,
     administrators, and other users.   Its contents are displayed
     to each terminal user when he logs on the system, unless
     suppressed by the user.

buffer:   (1)  an area in main memory to or from which records are
     read or written.  Usually used in contrast with work area
     (q.v.), in which manipulations are performed.  By the use of
     buffers and a work area, the processing of a file's records
     by the CPU may be overlapped with the I/O operations which
     read or write them.  (2)  to process a file using (usually
     two or more) buffers and a work area.

bug:   Computer slang for a program error or, more infrequently, a
     hardware error.

burst:    TO separate the sheets of fanfold paper at the

byte:   The contents of one IBM 360 or 370 memory location
     containing eight binary  digits; also, half of a halfword,
     one-fourth of a fullword, and one-eighth of a doubleword;
     contains one character in the EBCDIC coding scheme.

card:   A rectangular piece of light cardboard or heavy paper,
     approximately 3 inches by 7 inches, in which information may
     be recorded by means of  punched holes in predetermined
     locations.  Cards may contain up to 80 bytes of information.

NCC-IBH USER'S GUIDE        GLOSSARY                     JAN 1988

card punch:    A device which punches information into cards.
     Typically the speed of a card punch is between 100 and 600
     cards per minute.

card reader:    A device which reads the information punched in
     cards.   Typical card reader speeds are between 300 and 1000
     cards per minute.

catalog :    A file of data set names containing information about
     each data set, including its location.

cataloged procedure:   A set of OS Job Control Language (JCL)
     statements that can be invoked to perform a specific
     predefined function by an execute (EXEC) statement much as
     one executes a program.

cathode ray tube (CRT): A picture tube that creates visual displays
     using a beam of electrons illuminating an emulsion on its
     face .

CCW :   From Channel Command Word.  A doubleword in main memory,
     which contains a command for the control of a device.  CCW's
     are executed by the channels, device controllers, and devices
     asynchronously with CPU instructions.

channel:    A device that directs information flow between the
     central processing unit and input/output devices .

checkpoint:    A reference point in a program at which information
     about the contents of core storage can be recorded so that,
     if necessary, the program's execution may be restarted at an
     intermediate point.

class:    A collection of RACF-defined entities with similar

class authority:   An authority that allows a user to define
     entities to RACF in the classes defined in the class
     descriptor table.

class name:    The name that identifies a RACF class of entities.
     The class names are USER, GROUP, DATASET, and those class
     names found in the class descriptor table.

COBOL:    From Common Business Oriented Language.  A programming
     language whose verbs and syntax are based on English, used
     for data handling applications rather than scientific uses.
     An international standard exists which fully defines a
     standard COBOL language.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

cold start:   The restarting of a computer system from a normal or
     abnormal inactive condition, characterized by the loss of all
     pending operations, programs, and jobs.

COM:   Computer Output Microfilm, an output medium where
     characters and/or graphic images are formed on photographic
     film and then chemically developed for use.

command name:   The first term in a command, usually followed by

command processing:   The reading, analyzing, and performing of
     commands issued via a console or through an input stream.

common area:   The area of virtual storage that is addressable by
     all address spaces.

compiler:   A processing program that translates a high-level
     language (such as COBOL, FORTRAN, etc.) into a machine
     language object program.

completion code:   A number returned by a program or the system to
     its caller which expresses the success or failure of the
     attempted operation.  This usually refers to an abnormal
     termination code, issued by the system (System Completion
     Code) or internal to the program using an explicit ABEND
     macro (User Completion Code).

concatenate:   To arrange contiguously.

concatenated data sets:    A group of separate data sets which are
     treated as a single data set for the duration of a single job
     step .

concentrator:   A data communications device that relays messages
     from multiple, usually single-terminal lines, into a single
     high-speed trunk line.

condition code;   Code returned by a program to reflect certain
     conditions at termination.  Can be used in subsequent job steps
     which will be executed according to conditional tests on this
     code.  By convention in IBM 360 or  370 systems, the completion
     code is zero or a positive integer less than 4096, and the
     smaller the value,  the greater the degree of success indicated.

connect profile:   A description of a RACJ'-defined user's
     relationship to a group including group, authority and
     group-related user attributes.

conversational:   At the NCC, denotes a job which interacts
     with a low-speed terminal during its execution (of. batch).


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

core dump:    see dump (2).

cps:    From Characters Per Second, and synonymous therewith.
     Typically, transfer rates for high-speed devices (tapes and
     disks) are expressed in characters per second.

CPU:    From Central Processing Unit.  That portion of a computer
     which contains the logic circuits, internal registers, and
     frequently the main memory.
cross-reference listing:   A table, usually produced by a
     compiler or assembler, of the variable and/or instruction
     labels in a program and the instructions which reference

CRT:    See Cathode Ray Tube.

current connect group:   The group to which a user is associated
     during a terminal session or batch job.

CUU:    An acronym for Computer Utilization Unit.  A CUU is a
     small unit of work performed on the NCC-IBM computer system.
     It may be composed of central processor usage (CPU time),
     memory usage, I/O operations, or a combination of some or all
     of those items.  For details on CUU computation, see Section

cylinder:   A series of tracks in the same read/write head position
     on each disk surface forming a cylindrical arrangement.
     All tracks accessible without moving the read/write heads.

DAT:    Dynamic Address Translation.

data set:   A file of records, usually containing related data and
     having the same logical attributes.

data set control block (DSCB):   A record within the VTOC on a DASD
     volume describing the location and attributes of a data set on
     that volume.

data set name:   The character string which uniquely identifies to
     a computer system a particular data set on a specific volume.

data set profile:    A description of a RACF-defined DASD data set,
     including data set name, volume serial number, universal
     access, owner.

data base:    A collection of data used for information retrieval
     and reporting, usually a collection of data sets.

NCC-IBM USER'S GUIDE        GLOSSARY                     OAN 1988

DCB=    From Data Control Block.  An area of memory, jointly
     maintained by a program and the operating system, in which is
     recorded information concerning the attributes and current
     status of a data set.

DD=    From Data Definition.   A type of JCL card used to describe
     the device assignment and attributes of a data set.

debug:    To remove the "bugs" from a program, system,  or device.
     Most frequently used to denote an organized attempt to detect
     and correct all of the  errors which exist by systematic and
     repetitive review, testing, and correction cycles.

decollate:   To separate the parts of multipart printed output,
     usually by means of a machine designed for the purpose.

default group:   The group to which a user is associated when a
     group name is not specified on the TSO LOGON command or batch
     JOB statement.

default value;   The choice  among exclusive alternatives made by
     the system when no explicit choice is specified by the user.

degausser:   A device which  demagnetizes an entire reel of
     computer tape,  thereby  erasing any data contained thereon.

delimiter:   Bound.   For instance, many character strings are
     bounded by special characters such as quotes which are
     referred to as  delimiters.

demand  paging:   Transfer of a page from external page storage to
     real storage at the time it is needed for execution.

density of reference:   The  attribute whereby most of  a page's
     instructions and/or data are used while that page is in
     real storage.

direct  access:   Characterizing an access method or data set in
     which records are read  or written in an arbitrary order.  All
     direct access data sets must reside on disk volumes.

disk:    A high-speed storage medium which consists of  a number of
     disks mounted on a single vertical spindle rotating past a set
     of read/write heads in  such a fashion that very rapid access is
     possible to any data location.

doubleword:   Eight  contiguous bytes in main memory, aligned so
     that the address of the leftmost byte is divisible by eight.

down;    Inoperative, especially when applied to the hardware or
     software of a computer  system.


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

DSCB=  See Data Set Control Block.

DSN=   See Data Set Name.

dump:   (1)  TO copy, especially data from an on-line volume or
     medium to an off-line volume or medium.  (2)  To copy the
     contents of all or part of a computer's main memory to a
     printer; also, the printed result of such an operation.

dynamic address translation (DAT):   (1)  The change of a virtual
     storage address to a real storage address during the
     execution of an instruction.  (2)  A hardware feature that
     performs the translation.

dynamic allocation:   Assignment of system resources to a program
     at the time the program is executed rather than at the time
     it is loaded into main storage.

dynamic storage allocation:   A storage allocation technique in
     which the location of computer programs and data is
     determined by criteria applied at the moment of need.

EBCDIC:   From Extended Binary Coded Decimal Interchange Code.  A
     scheme for encoding characters as arrangements of eight bits.
     This is the coding scheme used internally in IBM  360 and
     370 computers .

EC mode:   Extended control mode.

EOB:   End of block.

EOD;   End of data, usually the end of a data set.

EOF:   End of file, usually the end of a data set.

EOT:   End of tape.

       End of volume.
EXCP :    From execute channel program.  Denotes a machine-language
     command used to initiate all I/O operations.  Typically an
     EXCP is required for each physical record read or written,
     but a single EXCP may control many such operations.

EXEC:    From execute.  A JCL card used to initiate the execution
     of a program or cataloged procedure.

extended control (EC) mode:   A mode in which all the features of
     a System 370 computing system, including dynamic address
     translation, are operational.  See also Basic Control (BC)

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

     mode .

external page storage:   The portion of auxiliary storage that is
     used to contain pages.

file:   Synonymous with data set.

fixed length:   a record format in which every record must contain
     the same number of characters.   An indication of the record
     length may or may not be included in each record, but none is
     required by OS.

fixed page =    A page in real storage that is ineligible for
     page-out while the address space owning that page occupies
     real storage.  This page (and any others belonging to that
     address space) may be paged out if the entire address space
     is swapped out of real storage.

foreground:    (1)  jn multiprogramming, the environment in which
     high-priority programs are executed.  (2)  Under ISO, the
     environment in which programs are swapped in and out of main
     storage to allow CPU time to be shared among terminal users.
     All command processor programs  execute in the foreground.
     Contrast with Background.

foreground  job:   (1)  A high-priority job, usually a real-time
     job.  (2)  A teleprocessing or  graphic display job that has
     an indefinite running time during which communication is
     established with one or more users at local or remote
     terminals.  (3)  Under TSO, any job executing in a swapped
     region of main storage, such as a command processor or a
     terminal user's program.  Contrast with Background Job.

foreign:   At the NCC-IBM, characterizing a tape volume which is
     not controlled by the THS.

FORTRAN:   From Formula Translation.  A procedure-oriented
     programming language, primarily for computational

fragmentation:   The ability to assign real storage locations to
     virtual addresses because the available spaces are smaller
     than the page size.

frame:    one bit from each of the longitudinal tracks on a reel of
     magnetic computer tape.

fulluord:   Four contiguous bytes in main memory, aligned so that
     the address of the leftmost byte is divisible by four.

generation  data group:   A group of  data sets, kept in


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN  1988

     chronological order.  Any of these data sets may be precisely
     specified by the name of the group and a relative generation
     number, where zero is the number of the most recent data set,
     minus one is the number of the next most recent, and so

generation data set:   One of the data sets in a generation data

group:    A collection of RACF users who can share access
     authorities for protected resources (equivalent to user

group authority:   An authority that relates to a type of function
     a user can perform in a group.  The group authorities are

group data set:    A DASD data set defined to RACF where the
     first-level qualifier of the data set name is "GP".   The
     data set is owned by the account specified as the second-level

group name:   one to eight alphanumeric characters beginning with an
     alphanumeric, #, S, or 
NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

housekeeping:    Often used to refer to that section of a program
     which is  executed once at the beginning of processing and not
     thereafter.  Usually this section contains instructions to
     initialize variables, parse and act on input parameters, open
     files, etc.

IBM - International Business Machine Corporation, a
     manufacturer of large-scale computers.

indexed sequential:   An IBM access method in which records may be
     retrieved either in order by a key or randomly.

initials:    At the NCC, the three characters which, when appended
     to the characters "EPA", form a user's User-ID.

initiator/terminator:   The job scheduler function that selects
     jobs  and  job steps to be executed, allocates input/output
     devices for them, places them under task control, and at
     completion of the job supplies control information for
     writing job output on a system output unit.

input:    The transfer of data into a computer system's main
     memory, or the data so transferred.

input stream:    A collection of control statements and input data
     submitted to the operating system on a device, such as a
     card  reader, especially activated for this purpose by an

instruction:   The smallest unit of work which a computer may

interactive:   A synonym for conversational (q.v.).

internal reader:   A facility that transfers jobs to the job entry
     subsystem (JES2).

internal writer;   A facility in the job entry subsystem (JES2)
     that  allows user written output writers to write data on
     devices not directly supported by the job entry subsystem.

interpreter:   A language processor which operates on
     original-form source statements held in memory,
     "interpreting" them as they are to be executed.  This process
     is considerably more expensive than the execution of a
     compiler's output, which consists of machine-executable
     instructions generated from the original source statements.

interrupt-1   A signal generated in some part of a computer system
     which indicates the occurrence of a particular event.  Such
     an event  may be internally generated, such as the completion


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

     of a previously issued command to a device, or externally
     generated, such as the arrival of a job in the hopper of a
     card reader which is signaled to the computer system by the
     operator's depression of the reader's start button.  Under
     OS, such signals cause the problem program currently
     operating to be temporarily suspended while the system reacts
     to the signal (handles the interrupt).  Accordingly systems
     such as OS are often referred to as interrupt-driven.

1/0=   Input or output, or input and output.  Often the term
     is used to refer to certain device operations which do not
     actually involve the transfer of any data  (any input or
     output) as well as to those operations which do involve such

IPL=   From Initial Program Load.  The process  of loading the
     operating system into an inactive computer and initiating its
     processing of jobs.

JCL:   From Job Control Language.  The descriptive and control
     statements used to direct an operating system in the
     processing of jobs; more specifically, the statements used
     to direct OS in the processing of jobs in  an IBM 360 or 370
     computer system.

JES2:   A program that handles the order of job execution,  job
     submission, and SYSOUT (printed and punched) output
     processing.  The program controls the communication with high-
     speed Remote Job Entry workstations, handles commands (e.g.,
     from WYLBUR, ISO, workstations) related to job and system
     status, and handles FETCH supplying job output to WYLBUR.   It
     provides for the routing of jobs to be input, executed, and
     their data sets to be printed/punched on any subsystem in a
     multisubsystem environment.  It is the OS/MVS parallel
     software to HASP in OS/MVT.

JES reader:   In OS/VS, the part of the job entry subsystem that
     controls the input stream and its associated job control
     statements.  Synonymous with output stream control.

JES writer:   in OS/VS, the part of the job entry subsystem that
     controls the output of specified data sets.  Synonymous with
     output stream control.

job:   An ordered sequence of job steps, usually interrelated.   In
     OS systems, a job is an indivisible unit as far as the
     control of work flow and accounting charges is concerned.

JOB:   A type of JCL statement which specifies  accounting and
     environmental information for an entire job.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

job control language:   see JCL.

JOBLIB:  In OS systems, the ddname given to those DD statements
     which describe the primary program libraries for an entire
     job .

job step:    A set of JCL statements which direct the operating
     system in the execution of a single program.  A job step may
     be conditionally executed based on the results of preceding
     job steps.

JSM:   From Job Stream Manager.  The programs and routines which
     direct the selection of jobs for processing.

K:    used after a number to denote 1000 bytes.  When denoting
     core, actually indicates  1024 bytes (MOO in hexadecimal),
     e.g.  1024K = 1 megabyte = ( 1024 )*( 1024) = (400 hex)**2 =
     1,048,576 actual bytes.  See megabyte.

keypunch;    A keyboard device for the manual entry of information
     in punched cards.

laser printer (3800):  High-speed, nonimpact printer using
     electrophotographic and laser technology; can print up to
     20,0040 lines per minute; can be used in on-line or off-line
     mode .

library;   Often used as a synonym for partitioned data set

line printer:   A device which prints a line at a time.  Typically,
     the speed of a line printer is between 300 and 2000 lines per

linkage editor:    ft program provided by IBM as an adjunct to the
     various language translation programs.  The Linkage Editor is
     used to process the object decks produced by the language
     translators, resolving references between programs, changing
     the form to that of a relocatable load module, and storing
     attribute information.

loader:    A program which reads load modules into the main
     memory of a computer system, thus preparing them for execution

load module:   The form of a program which may be loaded directly
     into a computer's main memory and then executed.  A load
     module may be fully link-edited (with all external references
     resolved) or partially link-edited (with at least some
     external references still to be resolved).

locality of reference:    An attribute of a program whose


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN  1988

     references to instructions and/or data are confined to as feu
     pages as possible for as long as possible.

logical record:   A single record of a file, without regard to the
     conventions used in reading or writing it.  Logical records
     may be read or written alone (unblocked) or as one of a  group
     of similar records (blocked) on the storage medium.

main storage:   (1)  The general purpose storage of a computer.
     Usually, main storage can be accessed directly by the
     operating registers.   Contrast with auxiliary storage.   (2)
     All program-addressable storage from which instructions  may
     be executed and from which data can be loaded directly into
     registers .

master address space;   The virtual storage used by the master
     scheduler task.

megabyte:     denoted by M; e.g. 1M is one megabyte
     (1,048,576 actual bytes).  A register has 3 bytes (24 bits)
     available for addressing giving an addressing capability
     of 2**24=16**6=16M=16,777,216 bytes=1,000,000 hex=16384K.
     See K.

memory:   That portion of a computer system which is designed to
     accept and store information or CPU instructions in such a
     way that they are available for later recall.  Typically, a
     computer's main memory is arranged serially and data items
     may be stored or retrieved in any sequence, the specification
     of the memory location being by its address (that is, its
     position relative to a fixed starting point).

memory dump:    See Dump (2).

memory speed:   The rate at which characters can be read from or
     written to a computer's memory.  Usually, this rate is
     expressed as a cycle time, the time required to read or  write
     the smallest possible group of characters.  Typically, cycle
     times for large-scale computers lie between several hundred
     nanoseconds and several microseconds for from four to sixteen

microsecond:    One millionth of one second.

migration:   The period of time during which an application system
     is transferred to a new operating environment.

millisecond:    One thousandth of one second.

multiple address space:   A feature that provides each use with a
     private address space.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988
multiprocessing:   The process of allowing more than one program
     to execute concurrently in a single computer system which
     contains more than one CPU.  In a multiprocessing system,
     the instruction processing for more than one program may be
     proceeding simultaneously in different CPU's.

multiprogramming:   The process of allowing more than one program
     to execute concurrently in a single computer system.  In a
     multiprogramming system, the instruction processing (CPU
     time) of one program is performed while I/O operations for
     other programs are being executed by the channels or by I/O
     processors .
       see os/nvs.

MVT:   See OS/MVT.

nanosecond:   one billionth of a second.

national character:    One of the special characters at-sign ( 
NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

     Basically, virtual memory involves the use of high-
     speed,  direct-access storage devices as secondary memory.
     Special hardware and software supervise the "paging in" and
     "paging out" of programs between main memory and the
     secondary or virtual memory.

OS/MVT:   Operating System/Multiprogramming with a Variable
     number  of Tasks.

OS/VS:    See OS/MVS.

OS/VS2:   See OS/J1VS.

output:   The transfer of data out of a computer system's main
     memory, or the data so transferred.

overlay:   a segment of a program or program-collection which,
     when active (being executed), occupies the memory otherwise
     occupied by some other inactive segment of the same program
     or program-collection.  Often used as an adjective describing
     a program which employs this technique.

page:   (1)   A fixed-length block of instructions, data, or both
     that can be transferred between real storage and external
     page storage.   Note that page and pt^a frame are not
     synonymous.  It is important to preserve the distinction
     between a page -- the data which is transferred between real
     and external page storage -- and the frame or slot in which
     it can  reside.  (2)  To transfer instructions, data, or both
     between real storage and external storage.

page data set:   a data set in external page storage, in which
     pages are stored.

page fault;    A program interruption that occurs when a program
     references an instruction or data in a page that does not
     currently reside in main storage.

page frame:    a block of real storage that can contain a page.
     Synonymous with frame.

page-in:   The process of transferring a page from external page
     storage to real storage.

page-out:   The process of transferring a page from real storage
     to external page storage.

paging:   The process of transferring pages between real storage
     and external page storage.

paging device:   A direct access storage device on which pages

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

     (and possibly other data) are stored.

parameter:    An input argument to process.  Usually the value of a
     parameter exerts a significant control on the process and is
     assigned externally to the process.

partitioned data set:   The combination of a collection of
     sequential files (the members of the partitioned data set)
     and a directory to the location of each, all in a single data
     set on a direct-access volume.

password;   At the NCC, the character-string associated with a
     User-ID; by the proper entry of a User-ID and password
     together the computer system can confirm the identify of any
     user.   Theoretically, the password for any User-ID should be
     known only to the corresponding user.

PDS;    From Partitioned Data Set, and synonymous therewith.

physical record:   The record or group of records which is written
     to and read from the storage medium of a data set as a single
     entity.   For blocked files, each block is a physical record;
     for unblocked files, each logical record is a physical

plotter:    A  device which draws lines on paper under program

PL/1:    A high-level programming language which contains, in a
     single language, many of the features and attributes of
     COBOL, FORTRAN, and Assembler language.  PL/1 was developed
     by IBM.

PM:    From Preventive Maintenance.  The periodic scheduled
     maintenance of a piece of equipment, especially of the
     hardware of a computer system.

printer:    Frequently used as a synonym ^.-. r line printer.

print train:    On some printers, a replaceable cartridge which
     contains the character set to be used.   The use of different
     print trains at different times allows different character
     sets to  be used on the same printer.

priority:   An attribute of a job, usually a number, which
     expresses an expected service time and frequently a cost
     multiplier.  By convention at the NCC,  the higher the
     priority of a job, the shorter its targeted  service time and
     the higher its cost.

private address space:   An address space assigned to a particular


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

     user .

procedure:    Frequently used as a synonym for cataloged procedure.

processor storage:   General purpose storage that is part of a
     central processing unit.  Synonymous with real storage.

profile:   A description of the characteristics of a RACF-defined
     entity.  A profile resides on the RACF data set.  See also
     connect profile, data set profile, group profile, user

program:   An organized collection of instructions which direct a
     computer system in the execution of a particular task.

protected resource:    A resource that is defined to RACF for the
     purpose of controlling access to the resource.  With RACF,
     the resources that can be protected are DASD data sets, tape
     volumes,  DASD volumes, terminals, and any other resources
     defined in the class descriptor table.

PTF:   From Program Temporary Fix.  A modification to a program,
     usually to correct an error, provided in a temporary form
     rather than as a replacement program.

punch:   Frequently used as a synonym for card punch.

queue:   An ordered list of entities waiting (queued) for some
     resource or service.  The items in a queue may be jobs,
     programs, I/O operations, or any of many other subdivisions
     of a computer's worK .

RACF:   See resource access control facility.

random access:   Access in an arbitrary order, especially to the
     records in a data set.

reader:   Frequently used as a synonym for card reader.

real address:    The address of a location in real storage.

real storage:    The storage of a System 370 computing system from
     which the central processing unit can directly obtain
     instructions and data, and to which it can directly return

real storage management-1   Routines that control the allocation of
     pages in real storage.  Abbreviated RSM.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAK 1988

record form or record format:   A designation of the
     characteristics and physical arrangement of the records in a
     data set recognized by the operating system.  At the NCC ,
     the permissible record formats are F (fixed-length), V
     (variable-length), or U (undefined).  Records which are F or
     V may be blocked (FB or VB), and records which are V may
     be blocked and spanned (VBS).

recovery routine:    a routine that is entered when an error occurs
     during the performance of an associated operation.  It
     isolates the  error, assesses the extent of the error,
     indicates subsequent action, and attempts to correct the
     error and resume operation.

recovery termination manager:   A program that handles all normal
     and abnormal  termination of tasks by passing control to a
     recovery routine associated with the terminated function.

register:   A location in a CPU in which information may be stored
     independently of the main memory.  In most computer systems,
     some registers are accessible to the programmer and others
     are not.  In  reference to the IBM computer at the NCC,
     register usually refers to one of the sixteen "general
     purpose registers" or to one of the four "floating-point
     registers".  A register has 2M bits which can be used for
     addressing giving an addressing capability up to 16
     megabytes.  See megabyte.

resident;   Pertaining to a program that is permanently located in
     storage.  For example, the nucleus in main storage or a
     system library on direct access storage.

resource:   A facility of the computing system or operating system
     required by  a job or task, and including main storage,
     input/output  devices, the central processing unit,
     data sets, and control or processing programs.

resource access control facility (RACF):   An IBM program product
     that can help a data processing installation limit access
     to its resources by controlling users (or who may use the
     system), resources (or what parts of the systems each user
     may access),  and access authorities (or how each user may
     access those  resources).

ROE:    From Remote Job Entry.  The process of submitting jobs to a
     computer system at a site physically distant from it but
     connected via communications lines.

secondary storage:   Auxiliary storage.

security:   Prevention of access to or use of data or programs


NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

     without authorization..

seek:   The movement of a read/write head on a direct access
     device from one cylinder to another; less frequently such a
     movement from one track to another.

segment:   A contiguous 64K area of virtual storage.

sequential:   A data set organization characterized by the
     necessity for reading or writing the records of a file only
     in their physical sequence.

SMF :    From System Management Facility.  A collection of routines
     in the operating system which collect statistics on the
     execution of jobs for billing or analysis purposes.

snap or snapshot:   A dump, usually of only a small portion of
     memory, which is taken by request rather than due to an

software:   The programs used on a computer system.

sort:   (1)  TO arrange a group of items in ascending or
     descending sequence by one or more keys.  Typically "sort" is
     used to refer to such a process where the input and output
     reside on an external medium and the term "sequence in
     memory" is used to describe the sequencing of items in a
     table or array wholly contained in main memory.  (2)  A
     program which accomplishes the sorting function described

source:   The instructions of a program as the programmer wrote
     them, i.e., before translation into internal form.

spanned:   A record format in which the logical records may be
     longer than the physical blocks; hence records may span two
     or more blocks.

special character:   A character, usually a printable character,
     which is neither a letter nor a numeric digit.  A blank
     may or may not be considered a special character,
     depending on the situation.

spooled data set:   A data set written on an auxiliary storage

spooling:   The process of storing card images or print lines on a
     high-speed device (tape or disk) so that a program's
     execution is not slowed by a dependence on the very slow
     speed of a reader, punch, or printer.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN  1988

STEPLIB:   In OS systems, the ddname given to those DD statements
     which describe the primary program library for a job step.

strip:   TO remove the pinfeed margins from printed computer

subroutine:   A program or program-section which performs a
     specific, narrowly defined function.  Usually subroutine
     refers to a relatively small program.

swap:   (1)  in systems with timesharing, to write the main
     storage  of a job auxiliary storage and read the image of a
     job into main storage.  (2)  Under OS/VS2 with ISO, to write
     the active pages of a job to external page storage and read
     pages of another job from external page storage into real

swap data set:   A data set dedicated to the swapping operation.

system management facilities (SMF):   An optional control program
     feature  of OS 360 and OS/VS that provides the means for
     gathering and recording information that can be used to
     evaluate system usage.

tape:   A high-speed storage medium which consists of a narrow
     strip of magnetically sensitive tape passed under a
     read/write head.  Tape data storage is characterized by
     sequential access.

terminal:   A low-speed keyboard device used for human-to-computer

thrashing:   A condition to which the system can do little useful
     work because of the excessive paging.

IMS:   From Tape Management System.  A software package used at
     the NCC  to control the creation, use, and scratching of
     magnetic tape volumes.

track:   The  intersection of a cylinder on a disk with one of the
     recording surfaces.

undefined or  undefined-length records:    A record format in which
     the number of characters in each record of a file may vary
     (but there is a maximum length).  Each record may or may not
     contain  an indication of the length of the record,  but no
     such indication is required by OS.

unit record device:    Devices such as readers and punches that
     deal with card images.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

universal access:    The default access authority that applies to a
     resource if the user or group is not specifically permitted
     access to the resource.  Universal access can be any of the
     access authorities.

up:    Fully operational and available, especially when applied to
     a computer system or one of its components.

user attribute:   A characteristic of a user that defines the type
     of functions  the user can perform on entities.   The user
     attributes are GRPACC, ADSP, and REVOKE.

user data set:   A DASD set defined to RACF where the second-level
     qualifier of  the data set is a User-ID.

User-ID:   At the  NCC, a six-character code which uniquely
     identifies a  particular individual authorized to use the
     computer system.  Each User-ID consists of the characters
     'EPA1 followed by three letters which are often referred to
     as that user's initials.

user profile:  A description of a RACF-defined user including
     User-ID, user name,  default group name, password, owner,
     access authority and attributes.

variable-length records :    A record format in which the number of
     characters in each record of a file may vary (but there is a
     maximum length).  OS requires that each variable-length
     record be prefixed by a 'Record Descriptor Word' or RDW which
     contains the  length of the record.

VIO:   Virtual I/O, a system operation that transfers data using
     the system paging routines rather than actually reading from
     or writing to a real I/O device.

virtual address:   An address that refers to virtual storage and
     must, therefore, be translated into a real storage address
     when used.

virtual storage:   Addressable space that appears to the user as
     real storage, from which instructions and data are mapped
     into real storage locations.  The size of virtual storage is
     limited by the addressing scheme of the computing system and
     by the amount of auxiliary storage available, rather than by
     the actual number of real storage locations.

virtual storage management (VSM):   Routines that allocate address
     spaces and virtual storage areas within address spaces, and
     keep a record of free and allocated storage within each
     address space.

NCC-IBM USER'S GUIDE        GLOSSARY                     JAN 1988

volume:   A unit of mass storage characterized by a single unique
     identity to the operating system.  Frequently, but not always,
     a volume is physically separable from all other volumes.

volume serial number:    At the NCC, the six-character code (not
     necessarily numeric) which identifies a particular volume.

VS:   See OS/MVS.

VS2:   See OS/MVS.

VTOC=   From Volume Table of Contents.  A special data set on a
     direct access  volume which contains information describing
     each data set  on that volume.

warm start:   The  restarting of a computer system from a normal or
     abnormal inactive condition, characterized by the
     preservation  of all pending operations, programs, and jobs.

word:   (1)  That  group of characters in a computer system which
     is internally  accessed as a group by the hardware.  (2)  At
     the NCC, synonymous with fullword.

work area:   An area in main memory in which manipulations are to
     be performed  on the records in a file (cf. buffer).

working set'-   The  specific pages referenced during program
     execution over some arbitrary interval of time.  A program
     does not have  a single working set, but a series of working
     sets depending on the time the measurement starts and the
     duration of the interval.

NCC-IBM USER'S GUIDE          INDEX                      JAN 1988
* commands	3-20, D-3
$$LOGON	3-52
*AI	3-23
*AR	3-21
*BI	3-25
*RA	3-24
*RB	3-25
*US	3-30
                         — A --
Account Codes
   on JOB statement	4-2
ADP Utilization Identifier	•	3-5, 4-1
Advanced Systems, Inc.
   courses	2-20
AFTER control statement	3-17, C-1
Alien tapes	3-37
Allocating files	4-26
Application packages
   data base	7-2
   econometrics	7-10
   FOCUS	7-26
   mathematical	7-11
   plotting and graphics	7-13
   retrieval and reporting	7-5
   simulation	7-24
   statistical	7-8
   data set	3-21
   available at NCC	5-3
                         — B —
   procedures	3-8
   interpreter	6-15
   publications	6-15
Batch mode
   password	3-7
Batch processing	5-44
Battelle's Automated Search Information System...see BASIS

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

BEFORE control statement	C-2
Billing	2-24
Bin Access Authorization	4-55
Biomedical programs	7-8
BITNET Membership	2-11
   on DD statement	4-13, 4-24
   space allocation	3-32, 4-17
BMDP	see Biomedical
Broadcast messages	2-14
Building Indexes
   System Catalog	3-12
Bulk Data Transfer	5-11
Bursting	4-62
Business Hours	2-12
                         — C --
CalComp	2-5
   costs	2-28
   software	7-19
CAM	see Cartographic
Cancelled jobs	2-17
Card parameter
   on JOB statement	4-3
   on JOBPARM statement	C-4
   costs	2-28
Cartographic Automatic Mapping	7-20
   System	3-12
Catalogued procedures	4-5, 4-35
   adding DD statements	4-39
   application packages	7-1
   Assembler	6-3
   BMDP	7-9
   COBOL	6-7
   DC2	7-23
   defined	4-33
   ESP	7-11
   FORTRAN	6-11
   GPSS	7-25
   IMSL	7-12
   library	4-33
   naming	4-43

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

   overriding	4-5, 4-37
   PANVALET	7-22
   PL/1	6-13
   PROC statement	4-35
   sample	4-39
   SAS	7-10
   SIMPL/1	7-25
   symbolic parameters	4-35
   user	4-42
Central data center	2-1
Central processor	2-8
Channels	2-2
Chargeback	2-24
   for services	2-24
Charges-see costs	2-24
CICS	3-55
   performance goals	3-55
   service level objectives	3-55
   of SYSOUT data sets	4-46
Classes	2-20
Cleaning tape	2-6
CLIST	3-52, 5-44
CNTL control statement
   catalogued procedures	6-7
   publications	6-8
Commerce , Department of	3-2
   available compilers	6-1
   COBOL	6-6
   FORTRAN	6-11
   PL/1	6-13
Compress	3-28
   availability	2-12
   configuration	2-2
   harduare	2-2
   mainframe description	2-2
   operating system	2-2
   performance	2-4
   power supply	2-5
   status phone	2-14
   UPS	2-5

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

Computer Based Training	2-20
Computer Output Microfilm	F-1
COND parameter
   on EXEC statement	4-6
   on JOB statement	4-4
   mainframe	2-2
Connect time
   costs	2-27
   SPACE	4-15
Control statements	5-14
   IEBCOPY	5-21
   IEHLIST	5-28
   IEHMOVE	5-24
   IEHPROGM	5-16
   JES2	C-1
   Linkage Editor	5-30, 5-31
Controllers	2-3
   operating	3-10
   VSAM	3-10
Copies parameter
   on JOB statement	4-3
   on JOBPARM statement	C-4
Costs	2-24
   batch job	2-25
   CalComp	2-28
   cards	2-28
   connect time	2-27
   dearchiving data set	2-28
   disk packs	2-27
   foreign tape	2-27
   interactive job	2-26
   keypunch	2-28
   micrographics	2-28
   miscellaneous services	2-27
   printer	2-27
   special delivery	2-28
   storage	2-27
   tape	2-27
   tape storage	2-27
   TSO	2-26
Courier Service	2-23
   on-line syllabus	2-20
   training	2-20
Customer Information Control System	see CICS

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

                         — D —
Daily statistics	5-44
   allocating space	3-32, 4-13
   Management	3-20
   shared	4-10
Data base management systems	7-2
   BASIS	7-14
   SYSTEM 2000	7-3
Data Base Security	3-8
Data Catalog 2	see DC2
Data Control Block	see DCB
Data Definition Statement	4-8
Data Facility Extended Function	see DFDF
Data Processing Support Services	2-21
Data set
   allocating space on disk	3-32, 4-13
   allocation	4-9
   archiving	3-21
   backup	3-27
   cataloging	4-9
   compress	3-28
   creation	4-9
   dearchiving costs	2-28
   exclusive control	4-10, 4-21
   group	5-42
   IEHPROGM	5-17
   Input only	4-26
   names	3-9, 3-10
   partitioned	5-42
   Read only	4-26
   reload	3-26
   renaming	5-16
   scratch	3-27
   scratching	5-16
   sequential	5-42
   shared control	4-10
   SYSOUT	4-45
   tape	4-19
Datagraphix laser printer	4-49
Datagraphix 9835 Laser Printer	2-8
Date and time formats	5-44
   on DD statement	4-12, 4-23
DC2	7-23
DD Name
   on DD statement	4-8, 4-10, 4-20

NCC-IBJ1 USER'S GUIDE          INDEX                      JAN  1988

DD statement	4-8
   overriding	4-5
   parameters	3-39, 4-9
Dearchive	3-24
   costs	2-28
Debugging and optimizing systems
   Optimizer III	7-23
Decollating	4-62
Degausser	2-6
   on DD statement	4-24
   tape data sets	4-24
Department of Commerce	3-2
Device-Driven Subroutine Libraries	7-21
   graphics	2-5, 2-8
   micrographics	F-1
   plotting	2-5, 2-8
DFEF	5-31
Direct Access Storage Device	see DASD
Directory Space
   partitioned data sets	4-15
Disaster Recovery	3-8
   costs	2-27
   data set	4-9
   data set allocation	4-9, 4-13
   devices	2-3
   functional specifications	2-4
   library	2-21
   management	3-20
   performance characteristics	2-4
   on DD statement	4-10, 4-20
   ADABAS	7-2
   Archive	3-21, 3-23
   backups	3-25
   BMDP	7-9
   Bulk Data Transfer	5-13
   catalogued procedures	4-34
   CUECHART	7-15
   data set	3-30
   DC2	7-23
   dearchive	3-24
   ESP	7-11

NCC-IBM USER'S GUIDE          INDEX                      JAN 1988

   FOCUS	7-27
   FORTRAN	6-11
   GEOMAP	7-17
   GPSS	7-25
   graphics	7-14
   IBM	2-13
   IDCAMS	5-18
   ISPF	3-52
   KWIC/KWOC	5-35
   Linkage Editor	5-29
   NCC	2-1M
   on-line	2-14
   PANVALET	7-22
   PL-MATH	7-13
   restore	3-26
   SCRIPT	7-26
   SIMPL/1	7-26
   space	3-28
   SURFACE II	7-18
   SYSTLM 2000	7-4
   TSO/E	3-49
   usage aids	2-14
   VSAM	5-32
DPSS	2-21
   devices	2-3
   Catalogued procedures	5-36
                         — E —
Econometrics, software
   ESP	7-10
Education Services	2-20
Electronic spreadsheet	7-32
   requesting	2-16
ESP	see Econometrics
Etime parameter
   on JOB statement	4-3
Exclusive control
   of data set	4-10, 4-21
EXCLUSIVE control statement	C-2
EXEC statement	4-5
   operands	4-5
   parameters	4-5

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

Execute statement	4-5
   of tapes	3-40
Expiration date
   tapes	3-40
                         — F —
Facility Impact Monitoring and Analysis System... see FIMAS
File management
   software packages	7-21
File management systems
   DC2	7-23
   PANVALET	7-22
File transfer	5-10
   Input Only	4-26
   Read Only	4-26
FOCUS	7-26
   interactive	5-45
Foreign tape
   costs	2-27
Foreign Tape Reports	3-46
Foreign Tapes	3-43
Form number	4-60
   for Datagraphix laser printer	4-49
   for IBM 3800	4-47
   preprinted	4-61
   special	4-3, 4-57
Forms parameter
   on JOB statement	4-3
   on JOBPARM statement	C-4
   values	4-57
   catalogued procedures	6-11
   compilation	6-11
   publications	6-12
Fourth Generation Language
   FOCUS	7-26
full-screen terminal	2-9
GDDM	see Graphical
                                                   Data Display
GDG	see Generation
                                                   data group

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

General Purpose Simulation System	see GPSS
Generation data group	5-17
Global Status Tracking
   nodes	5-9
GPSS	7-24
Graphical Data Display Manager	7-21
   CalComp	7-19
   CAM	7-20
   CUECHART	7-15
   Device Driven Subroutines	7-21
   devices	2-5
   DISSPLA	7-17
   GDDM	7-21
   GEOMAP	7-17
   InterOrg	7-16
   ISSCO-GKS	7-18
   output	4-62
   PLINIT	5-38
   POLYVRT	7-20
   SAS	7-17
   software	7-13
   SURFACE II	7-18
   Tektronix	7-19
   TELL-A-GRAF	7-15
   Utilities	5-36
Group data sets	5-42
                         — H —
Halon system	3-8
   NCC	2-2
   tape	3-35
   WIG	2-7
HELP Command	5-1
Hours of Availability	2-12
                         — I —
I/O Control	2-21
IBM documentation	2-13
IBM 3800
   Bursting output	4-62
   Description	2-7
   SYSOUT class	4-46
   Use of	4-46

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

IBM 3820 Page Printer	4-52
IBM 4381	2-7
ICF Catalogs	5-31
IDCAMS	5-14, 5-17
IEBCOPY	5-14, 5-20
   control statements	5-2 1
IEBGENER	5-14, 5-19
IEBISAM	5-14, 5-26
IEBUPDTE example	4-43
IEFBR14	5-14, 5-15
IEHLIST	5-14, 5-27
IEHMOVE	5-14, 5-23
IEHPROGM	5-14, 5-15,  5-17
   control statements	5-16
IGL	see Interactive
IMSL	7-11
In-stream procedures	4-35
   defined	4-34
   system status	2-14
   User Memos	2-14
Input only files	4-26
Input/Output devices	2-2
   TMS	3-4 1
Inquiry and Reporting System	7-5
Instruction	2-20
Integrated Catalog Facility	see ICF Catalogs
Intelligent Transparent Restore Facility	3-22
Interactive Graphics Library	7-19
Interactive processing	5-44
Interactive System Productivity Facility	see ISPF
Interagency Agreements	3-1
Internat'l Mathematical £ Statistical Library.... see IMSL
InterOrg	7-16
IRS	see Inquiry  and
ISPF	3-52
IXR	see Intelligent

JAN 1988
                         -- J --
JARS	2-24
   application packages	7-1
   compilers, assemblers, and interpreters	6-1
   DD statement	4-8
   EXEC statement	4-5
   JOB statement	4-1
   language processors	6-2
   micrographics	F-2
   OUTPUT JCL	4-27
   utilities	5-14
   control statements	4-3, C-1
      AFTER	C-1
      BEFORE	C-2
      CNTL	C-2
      JOBPARM	C-4
      MESSAGE	C-5
      OUTPUT	C-6
      PRINT	C-6
      PUNCH	C-6
      ROUTE	C-9
      SHARED	C-2
      SIGNOFF	C-10
      SIGNON	C-10
   defined	3-13
   functions	3-56
   operator commands	D- 1
      concerning a device	D-9
      concerning specific jobs	D-7
      MISC	D-12
      summary	D-3
JES2 global status tracking	5-9
   card sample	4-
   recurring	3- 7
   retrieving output	4-54, 4-56, 4-62
   Sample Jobstreams	2- 5
   Scheduling	3- 3
   statement	4-
   statement parameters	4-
Job control language	see JCL
Job Entry Subsystem	see JES2
Job queue
   display through JES2	D-1
Job Status Tracking System	5-10
Job Stream Manager	3-14

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

JOBLIB statement	4-7
JOBPARM control statement	C-4
Johnson Systems	see JARS
                         — K —
Key Word in Context	5-34
Key Word out of Context	5-34
   costs	2-23, 2-28
Keypunch Services	2-23
                         — L —
   on DD statement	4-24
   self-sticking	4-61
Laser printer	see IBM 3800
   Datagraphix 9835	2-8
   WIC	4-49
   Disk	2-21
   Tape	2-21
   private	3-52
Linect parameter
   on JOB statement	4-3
   on JOBPARM statement	C-4
Lines parameter
   on JOB statement	4-3
   on JOBPARM statement	C-5
Linkage Editor	5-14, 5-27,
                                                   5-28, 6-1
   examples	4-39, 5-31
Linklist	4-7
LMF-PC file transfer	5-10
Log parameter
   on JOB statement	4-3
Logon Procedures
   ISO	3-51
   on DD statement	4-13, 4-24

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

                         — M —
   output	M-54
Mailing address
   National Computer Center	2-2
   for reference	2-13
   ordering information	2-14
Mathematical systems	7-8
   IMSL	7-11
   PL-MATH	7-12
MegaCalc	7-32
Memory	2-2
MESSAGE control statement	C-5
   on-line aids	2-15
   Telecommunications	E- 1 , E-4
Microfiche	2-22, F-1
Microform	2-22 , F-1
Micrographics	F-1
   costs	2-28
   services	2-22
   systems	7-24
   on JOB statement	4-4
   on JOB statement	4-4
Mylar ribbons	4-59
                         — N --
Naming Conventions	3-9
   for data sets	3-9, 3-10
National Technical Information Service	3-2
NCC-IBM Users Guide
   maintenance	1-2
   obtaining	1-2
   updates	1-2
   accounts'	3-1
News Alerts	2-14
   on JOB statement	4-5
NTIS	see National

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

Nylon ribbons	4-57
                         — 0 —
Object code	6-1
Operations schedule	2-12
Operator commands	D- 1
   concerning a device	D-9
   concerning specific jobs	D-7
   HISC	D-12
   summary	D-3
Organization of NCC	2-1
   document quality print	4-59
   graphics	4-62
   labels	4-61
   retaining	4-56
   to Remote Stations	4-54
   Washington Information Center	2-2, 2-6
Output Processing	4-45
Over-the-counter service	2-2, 2-7
                         — p —
Paper	4-60
   on DD statement	4-9
   on EXEC statement	4-5
   on JOB statement	4-1
FARM parameter
   on EXEC statement	4-6
Partitioned data set	5-42
Partitioned data sets	4-16, 5-17
   directory space	4-15
   IEBCOPY	5-20
   IEBGENER	5-19
   IEHMOVE	5-23
   changing	3-7
   in-batch jobs	3-7
Passwords	3-6
   Changing in TSO	3-52
PC-LMF file transfer	5-10
Performance	2-4
performance goals	3-50
Phoenix courses	2-20

NCC-IBM USER'S GUIDE          INDEX                      JAN 1988

Physical Security	3-8
PL-MATH	7-12
   catalogued procedures	6-13
   compilation	6-13
   publications	6-15
   Anderson-Jacobson	7-21
   devices	2-5
   GDDM	7-21
   GEOMAP	7-17
   Houston Instruments	7-2 1
   initializing plot tapes	5-36
   output	14-62
   routing plots	5-36, 7-14
   software	7-13
   ZETA	7-21
PN chain	4-57
Positional parameter
   on JOB statement	4-1
Power supply	2-5
Preprinted forms	4-61
   document quality	4-52 , 4-59
   IBM 3800	4-48
PRINT control statement	C-6
Print trains	4-57
   costs	2-27
Printers	2-6
   classes	3-14
   PRTY on JOB statement	4-4
   requesting	3-14
Privacy Act	3-5
Private disk packs	3-30
   reporting	2-16
   tape	3-49
   Telecommunications	2-15, E-1
PROC	see Catalogued
                                                   see In-stream
Procedure Library-Mathematics	see PL-HATH

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

Procedure name
   on EXEC statement	4-5
Procedures	see Catalogued
                                                   see In-streara
                                                   see Private
Processor	2-2
Production Control	2-19
Programming assistance	2-15
protocol converter	2-9
PRTY parameter
   on JOB statement	4-4
Publications	see
   Assembler	6-6
   BASIC	6-15
   COBOL	6-8
   FORTRAN	6-12
   PL/1	6-15
PUNCH control statement	C-6
                         -- Q --
2ED	5-1
Sualified name
   defined	3-9
fiUEUE command	5-4
   JES2 status command	5-9
                         — R —
RACF	3-7 ,  3-10
   Read or input only files	4-26
RD parameter
   on EXEC statement	4-6
   on JOB statement	4-5
Read only files	4-26
Receiving files	5-10
   on DD statement	4-12, 4-23
   for cancelled jobs	2-17
REGION parameter
   on EXEC statement	4-6
   on JOB statement	4-4

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

   with NTIS	3-2
Registration Procedures	3-1
Releasing space	3-28
Reload	3-26
Remote Job Entry	3-55
   sharing IDs	3-56
Renaming data set	5-16
   Monthly Tape Usage	3-46
   problem	2-16
Resource Access Control Facility	see RACF
Restart parameter
   on EXEC statement	4-6
   on JOB statement	4-5
Retention of output	4-56
Retention of tapes
   EXPDT	3-40
Retrieval and reporting packages
   IRS	7-5
   Mylar	4-59
   nylon	4-59
   printing	4-59
      control statements	C- 1
      operator commands	D- 1
RJE Access	2-8, 3-55
   SPACE	4-15
Room parameter
   conventions	4-55
   SPACE	4-15
Route codes	4-55
ROUTE control statement	C-9
   Bin ID	4-55
   Plots	7-14
   Remote	4-54
   SYSOUT parameter	4-54

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

                         — S —
SAS	7-9, 7-17
   interactive	5-45
Schedule for NCC facilities	2-12
   job	3-13
Scratch tapes	3-40
Scratching data set	5-16
SDSF	see Spool
   Data base	3-8
   PAI data	3-5
   Physical	3-8
   System	3-6
Self-sticking labels	4-61
Sending files	5-10
Sequential data set	5-42
service level objectives
   CICS	3-55
   ISO	3-50
   consultation and assistance	2-15
   Cost	2-24
      Miscellaneous	2-27
   Courier	2-23
   Keypunch	2-23
   Micrographics	2-22
   Production Control	2-19
   schedule	2-12
Shared control
   of data set	4-10
SHARED control statement	C-2
Signing on
   JES2	C-1
SIGNOFF control statement	C-10
Signon Announcements	2-14
SIGNON control statement	C-10
SIMPL/1	7-25
   systems	7-24
Simulation and modeling systems
   GPSS	7-24
   SIMPL/1	7-25
Simulation PL/1	see SIMPL/1
Slot number	3-44

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

   descriptions	2-14
   summary	G- 1
   ISPF Editor	5-45
Space	3-28
   block allocation	4-14
   contiguous	4-16
   cylinder boundaries	4-16
   on DD statement	4-12, 4-13
   primary allocation	4-13
   releasing	4-15
   secondary allocation	4-13
   tracks/cylinders allocation	4-15
Special forms	4-57
Spool Display Support Facility	5-9
   MegaCalc	7-32
Standards and Practices
   application, packages	7-1
   catalogued procedures	4-33
   data set names	3-9
   Tape Data Storage	3-35
Statistical Analysis System	see SAS
Statistical systems
   BMDP	7-8
   ESP	7-10
   IMSL	7-11
   SAS	7-9
Status phone	2-14, A-1
STEPLIB statement	4-7
   Central	2-2
   disk	2-3
   drum	2-3
   tape	2-3, 3-35
Support Functions
   SYNCSORT	5-33
   of courses	2-20
Symbolic parameters
   coding	4-35
   general	4-34
   nullifying	4-36
   Documentation	5-34

NCC-IBM USER'S GUIDE          INDEX                      JAN  1988

SYSOUT Classes	4-45
SYSOUT data sets	4-45
System Catalog
   data set name requirements	3-9
   description	3-12
System linklist	4-7
System Security	3-6
System software	G- 1
System status phone	2-14, A-1
SYSTEM 2000	7-3
SYSUT1	5-30
                         — T --
   alien	3-37
   archive and dearchive	3-47
   Assignment	3-36
   backup	3-8
   cleaning	2-6, 3-49
   costs	2-27
   data sets	4-19
   degausser	2-6
   description	3-35
   devices	2-3
   disposition	3-46
   expiration	3-40
   foreign	3-43
   JCL Considerations	3-39
   label	4-24
   library	2-21
   maintenance	3-46, 3-49
   Management System	3-39
   Monthly Usage Report	3-46
   plotting	4-62
   recreating data sets	3-38
   retention period	3-38
   Sending Offsite	3-46
   slot number	3-44
   storage	3-35
   TAPESCAN	5-38
   temporary	3-38
   uninitialized	3-37
   work	3-38
   write ring	3-39

JAN 1988
Tape Accelerator	2-3
Tape Data Set
   sample	4-20
Tape Management Catalog	3-39
Tape Management System	3-39
   reports	3-46
Tapemark	3-44, 4-25
Tapes	3-45
   software	7-19
Telecommunication Service Request	2-10
   assistance	2-15, E-1
   network	2-8
   RJE Access	2-8, 3-55
Telephone Numbers	A-
   GEOMAP	7-7
   obtaining	2-9
Text Editors	7-26
Textual Information Management Systems
   BASIS	7-4
TIME parameter
   default value	4-4
   on EXEC statement	4-6
   on JOB statement	4-4
Timesharing	3-13, 3-49
Timesharing Option	3-49
TMC	3-39
TMS	see Tape
TMS Inquiries	3-41
TN chain	4-57
Today	5-44
Training	2-20
Transferring data
   Bulk Data Transfer	5-11
   through BITNET	2-11
   Accessing	3-51
   global status tracking	5-9
   HELP	5-1
   performance goals	3-50
   2ED	5-1
   2UEUE command	5-4

NCC-IBH USER'S GUIDE          INDEX                       JAN  1988

   Utilities	5-2,  5-4,  5-9
   vs. batch	5-44
   defined	3-49
TYMNET	2-9,  3-51, E-1
   on JOB statement	4-4
                         — U --
U.S. Department of Commerce	3-2
UID	3-5
Uninterruptible Power Supply	see UPS
   on DD statement	4-11,  4-21
   of NCC-IBM Users Guide	1-2
UPS	2-5
Usage aids	2-14
   Memos	2-14
   problem reporting	2-16
User Procedures	3-52,  4-42
User Support
   availability	2-12,  2-16
   services	2-16
User Support Department
   telephone numbers	A- 1
   obtaining	3-1
   on JOB statement	4-2
Users Guide
   updates	1-2
   IBM	5-14
   IEBCOPY	5-14,  5-20
   IEBGENER	5-14,  5-19
   IEBISAM	5-14,  5-26
   IEFBR14	5-14,  5-15
   IEHLIST	5-14,  5-27
   IEHMOVE	5-14,  5-23
   IEHPROGM	5-14,  5-15
   Linkage Editor	5-14,  5-28
   PLINIT	5-36
   BUEUE	5-4
   SDSF	5-9
   TAPESCAN	5-38
   TSO	5-1, 5-2, 5-4,

NCC-IBM USER'S GUIDE          INDEX                      JAN 1988

Utilization Identifier	3-5
                         — V --
Virtual Storage Access Method	see VSAM
   on DD statement	4-11
   of tapes	3-40
VSAM	5-32
                         — M —
Washington Information Center	2-9
   output	2-2
   output routing	4-54
Waterloo SCRIPT	7-26
WATS Numbers	2-9
World Data Bank	7-20
Write ring policy	3-39
                         — Y --
Yesterday	5-44
ZETA	7-2 1
                         -- 3 --
3270 terminal	2-9
3800 Laser printer	see IBM 3800

                        TABLE OF CONTENTS



    1.2.1    Numbering Systems  	  1-1
    1.2.2    Jobstreams	1-2
    1.2.3    Submitting Comments   	  1-2





    2.2.1    NCC Hardware	2-2    Processor  	  2-2    Central Storage  	  2-2    Input/Output 	  2-2
        2.2.1.M    Tape Accelerator 	  2-3    Controllers  	  2-3    Disk, Drum, and Tape  Devices  	  2-3    Plotting and Graphic  Devices  	  2-5    Auxiliary Devices   	  2-5    Computer Print and Card Read/Punch System   2-6

    2.2.2    WIC Hardware	2-7    Remote Job Control Configuration  	  2-7    Plotting and Graphic  Devices  	  2-8    Datagraphix 9835 Laser Printer  	  2-8


    2.3.1    Remote Batch Access   	  2-8
    2.3.2    Interactive Access 	  2-9
    2.3.3    Terminal Procurement  	  2-9
    2 . 3 . M    Telecommunication Service Request	2-10
    2.3.5    BITNET Membership  	   2-11

    2.4.1    Business Hours	2-
    2.4.2    User Information Services	2-    Availability of Manuals  	  2-    User Memos	2-    System Status Information  	  2-    Signon Announcements 	  2-    On-Line Usage Aids 	  2-14

    2.4.3    Consultation and Assistance  	  2-15    Telecommunications Assistance  	  2-15    User Support Services  	  2-16    Production Control Services  	  2-19    Education Services 	  2-20

    2.4.4    Data Processing Support Services (DPSS)  .   .  .  2-21    Tape Library	2-21    Disk Library	2-21    I/O Control	2-21    Micrographics Services 	  2-22    Keypunch Services  	  2-23    Courier Service  	  2-23

    2.4.5    Cost of Services	2-24    NCC Chargeback Algorithm 	  2-24    Charges for Computer-Related Services   .  2-27



    3.1.1    Registration for Accounting and Security .... 3-1
    3.1.2    NTIS Accounts	3-2
    3.1.3    User Identification Codes	3-2    Invalid User-ID  	 3-3    Invalid Account  	 3-4    Deleted User-ID or Deleted Account .... 3-4

    3.1.3    ADP Utilization Identifier 	 3-5


    3.2.1    Obtaining and Changing Passwords 	 3-6
    3.2.2    Resource Access Control Facility (RACF)  .... 3-7
    3.2.3    Data Bases	3-8

    3.2.4    Physical Security   	  3-8
    3.2.5    Data Security	3-8
    3.2.6    Disaster Recovery Plans   	  3-8


    3.3.1    Data Set Naming Conventions	3-9    Operating System  (OS) Standards   	  3-9    NCC-IBM Conventions for User Data  Sets  .   3-10    NCC-IBM Conventions for Naming VSAM  Data   3-10
                   Sets	3-10    NCC-IBM Conventions for System Data  Sets   3-11    System Catalog  	   3-12

    3.3.2    Job Scheduling	3-13    JES2	3-13    User Production   	   3-17

    3.3.3    Direct Access Storage Device  (DASD) Management   3-20    Archives 	   3-21    Intelligent Transparent Restore Facility   3-22    Reload From Archives  	   3-23    Incremental Backups   	   3-24    fiRELOAD's  	   3-26    Full Volume Saves   	   3-27    Scratch  	   3-27    Compress 	   3-28    Space Release   	   3-28   Acquiring Disk Space  	   3-28   Temporary On-Line Data  Set  	   3-30   Private Disk Packs  	   3-30   Efficient Use of  DASD at the NCC  ....   3-32

    3.3.4    Tape Management	3-35    Tape Data Storage	3-35    Tape Management System  	   3-39    Foreign Tapes   	   3-43    Magnetic Tape Maintenance and Disposition  3-46


    3.4.1    TSO/E	3-49    Logon Procedures  	   3-51    ISPF	3-52    **LOGON  	   3-52


3. 6
4. 0
4. 1

4. 2

U. 3

3.4.2 CICS 	
4.1.1 JOB Statement 	
4.1.2 Execute (EXEC) Statement Parameters JOBLIB and STEPLIB Considerations . . .
4.1.3 Data Definition (DD) Statement Disk Data Sets Allocating Direct Access Storage ....
Device (DASD) Space Tape Data Sets Read or Input Only Files 	
4.1.4 OUTPUT JCL Statement Parameters Implicit Mode Explicit Mode 	
4.2.1 Catalogued Procedures 	
4.2.2 In-Stream Procedures 	 	
4.2.3 Coding Symbolic Parameters 	
4.2.4 Nullifying Symbolic Parameters 	
4.2.5 Overriding Catalogued Procedures Overriding EXEC Statement Parameters . . Overriding DD Statement Parameters . . . Adding DD Statements to a Procedure . .
4.2.6 Sample Procedure 	
4.2.7 User Procedures 	
4.2.8 Private Procedure Libraries 	
4.3.1 SYSOUT Classes 	
4.3.2 Use of the 3800 Laser Printer 	
. 4-1
. 4-5
. 4-6
. 4-7
. 4-8
. 4-9

-------    Forms for the 3800	4-47
        4.3. 2. 2    Printing on the 3800 Printer	4-48    Job Control Language 	   4-49

    4.3.3    Use of the Laser Printers at WIC	4-49    Job Control Language 	   4-51    Forms for the WIC Las__  Printers  ....   4-51    Fonts   	   4-52

    4.3.4    IBM 3820 Page Printer	4-52
    4.3.5    Remote Routing	4-54
    4.3.6    Output Distribution and Room Parameters   .  .  .   4-54    Obtaining a Bin ID	4-55    Room Parameter  	   4-55

    4.3.7    Retention of Output in Print Queue  	   4-56
    4.3.8    Print Trains and Special  Forms  	   4-57
    4.3/9    Bursting and Decollating	4-62
    4.3.10   Plotter Output 	   4-62


5.1    TSO UTILITIES	5-1

    5.1.1    PRINTOFF	5-2
    5.1.2    2UEUE Command	5-4
    5.1.3    JES2 Global Status Tracking  	  5-9
    5.1.4    Spool Display Support Facility  	  5-9
    5.1.5    PC-LMF File Transfer	5-
    5.1.6    Job Status Tracking System  	   5-
    5.1.7    Bulk Data Transfer  (BDT)	5-    Interactive Execution   	   5-    Batch Execution   	   5-    BDT Features	5-    Documentation   	   5-13

5.2    IBM UTILITIES	5-14

    5.2.1    IEFBR14	5-15
    5.2.2    IEHPROGM	5-15
    5.2.3    IDCAMS	5-17
    5.2.4    IEBGENER	5-19
    5.2.5    IEBCOPY	5-20
    5.2.6    IEHMOVE	5-23
    5.2.7    IEBISAM	5-26
    5.2.8    IEHLIST	5-27
    5.2.9    Linkage Editor	5-28

-------    SYSLIN 	  5-29    SYSLIB 	  5-30    SYSUT1 	  5-30    SYSPRINT 	  5-30    SYSLMOD  	  5-30    INCLUDE Control Statement   	  5-31

    5.2.10   ICF Catalogs	5-31
    5.2.11   VSAM	5-32


    5.3.1    SYNCSORT	5-33
    5.3.2    KWIC/KWOC	5-34
    5.3.3    DSSUBMIT	5-36
    5.3.4    Initializing and Routing Plot Tapes	5-36
    5.3.5    TAPESCAN	5-38
    5.3.6    Using Full-Screen Facilities to Advantage   .  .  5-39    Stacking Commands  	  5-39    Using MYLISTC  	  5-39    Using MYED	5-40

    5.3.7    Using Group Data Sets	5-42
    5.3.8    Using Partitioned Data Sets	5-42
             vs. Sequential Data Sets	5-42
    5.3.9    Using Date and Time Formats	5-44
    5.3.10   Interactive Processing vs.  Batch  Processing   .  5-44   Advantages of Interactive Processing  .  .  5-44   Example of a CLIST	5-45



    6.1.1    Catalogued Procedures   	  6-3
    6.1.2    Publications	6-6

6.2    COBOL	6-6

    6.2.1    Catalogued Procedures   	  6-7
    6.2.2    Publications	6-8
    6.2.3    VS COBOL Migration	6-9

6.3    FORTRAN	6-11

    6.3.1    Catalogued Procedures   	  6-11
    6.3.2    Publications	6-12

6.4    PL/1	6-13

    6.4.1    Catalogued Procedures	6-13
    6.4.2    Publications	6-15

6.5    BASIC	6-15



    7.1.1    ADABAS Data Base Management System	7-2

    7.1.2    SYSTEM 2000	7-3     Publications  	  7-4     DBACCESS 	  7-4

    7.1.4    Battelle's Automated Search   	  7-4
             Information System  (BASIS)  	  7-4


    7.2.1    Inquiry and Reporting System  (IRS)  	  7-5
    7.2.2    EASYTRIEVE PLUS	7-7


    7.3.1    Biomedical Computer Programs  (BMDP)   	  7-8
    7.3.2    Statistical Analysis System (SAS)   	  7-9
    7.3.3    Econometric Software Package  (ESP)  	   7-10
    7.3.4    International Mathematical and  	   7-11
             Statistical Library (IMSL)  	   7-11     Catalogued Procedures   	   7-12     Publications  	   7-12

7.3.5    Procedure Library-Mathematics  (PL-MATH)   	   7-12


    7.4.1    TELL-A-GRAF	7-15
    7.4.2    CUECHART	7-15
    7.4.3    DATA CONNECTION	7-15
    7.4.4    TELLAPLAN EXPERT 	   7-16
    7.4.5    InterOrg	7-16
    7.4.6    DISSPLA	7-17
    7.4.7    GEOMAP	7-17
    7.4.8    SAS/GRAPH	7-17
    7.4.9    ISSCO-GKS	7-18
    7.4.10   SURFACE II Graphics  	   7-18
    7.4.11   Basic Tektronix Software  	   7-19
    7.4.12   Tektronix Interactive Graphics  Library  ....   7-19

    7.4.13   CalComp Software 	  7-19
    7.4.14   Cartographic Automatic Mapping 	  7-20
    7.4.15   POLYVRT  	  7-20
    7.4.16   Device-Driver Subroutine Libraries  	  7-21
    7.4.17   Graphical Data Display Manager (GDDN)   ....  7-21


    7.5.1    PANVALET Program Management   	  7-21
             and Security System	7-21
    7.5.2    Data Catalog 2 (DC2)	7-23



    7.7.1    General Purpose Simulation System (GPSS)  .  .  .  7-24
    7.7.2    Simulation PL/1 (SIMPL/1)  	7-25

7.8    TEXT EDITORS - SCRIPT	  7-26


    7.9.1    Capabilities of FOCUS	7-27
    7.9.2    Major Capabilities 	  7-27
    7.9.3    Extended Capabilities  	  7-28
    7.9.4    Security Capabilities  	  7-29
    7.9.5    Advantages and Disadvantages  	  7-29
    7.9.6    FOCUS Software Available on the IBM 3090  .  .  .  7-30
    7.9.7    Documentation	7-31














GLOSSARY      	      Glos-1

                         LIST OF FIGURES
Figure 2-1.

Figure 2-2.

Figure 2-3.

Figure 2-4.

Figure 3-1.

Figure 4-1.

Figure 4-2.

Figure 4-3.

Figure 4-4.

Figure 4-5.
Direct Access Storage Units   	  2-4

Types of Tape Drives	2-5

Printing Systems 	  2-7

Hours of Operation	2-13

Job Service Level Definitions   	   3-15

Data Set with 10,000 80-Byte  Records	4-17

Determining Block Size  	   4-19

7- and 9-Track Tapes	4-22

Values for DEN	4-24

Self-Stick Labels  	   4-62