\mi
U.S. Environmental Protection Agency
Office of Administration and Resources Management
Office of Information Resources Management
Washington, D.C. 20460
INFORMATION
RESOURCES
MANAGEMENT
IRECTIVES
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=
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
NATIONAL DATA PROCESSING DIVISION
NATIONAL COMPUTER CENTER
-------
Preface
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.
11
-------
NCC-IBM USER'S GUIDE JAN 1988
1 . 0 INTRODUCTION
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.
1.1 PURPOSE AND SCOPE
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 USING THE MANUAL
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.,
Section 3.3.3.2.
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
1-1
-------
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
applicable.
1.3 MAINTENANCE AND DISTRIBUTION PROCEDURES
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-
1-2
-------
NCC-IBM USER'S GUIDE JAN 1988
line data set JUSD.USERS.GUIDE(MANUAL). See Appendix H for
printing instructions.
1-3
-------
NCC-IBM USER'S GUIDE JAN 1988
2.0 GENERAL FACILITIES AND SERVICES
2.1 ORGANIZATION AND MANAGEMENT
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.
2.2 PHYSICAL FACILITIES
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.
2-1
-------
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.
2.2.1.1 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.
2.2.1.2 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.
2.2.1.3 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.
2-2
-------
NCC-IBM USER'S GUIDE JAN 1988
2.2.1.4 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.
2.2.1.5 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.
2.2.1.6 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
2-3
-------
NCC-IBM USER'S GUIDE
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
storage.
Model Number
and Unit
Description
STC 4000(6)
solid state
drum
STC
8650
STC 8380
L
Approximate
No. of Bytes
Per Unit
1 1 million
635 million
2 . 5 billion
Seek Time in
Milliseconds
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
2-M
-------
NCC-IBM USER'S GUIDE
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.
Unit
Description
Nine-track
1600/6250 bpi
Nine-track
800 bpi
Seven-track
200/556/800 bpi
L
Gap
Size
Inches
. 6/ . 3
. 6
.75
Tape
Speed
in . /sec
200
200
125
Start
Time
in millis
1.6/1.1
2 . 1
4 . 4
Stop
Time
>econds
.8/.S
1 . 0
2 . 2
1
Rewind Time
Full Reel
in Seconds
45
45
45
2.2.1.7
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.
2.2.1.8
Auxiliary Devices
The NCC installation includes a number of auxiliary devices
important to overall operations but not directly concerned with
data processing.
2-5
-------
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).
2.2.1.9 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.
2-6
-------
NCC-IBM USER'S GUIDE
JAH 1988
TYPE
CHAR SET
NUMBER
AVERAGE
SPEED
IBM 32 1 1
PRINTER
3216 prin-fc train
(2) T11 up/low
(2) PI 1 UCS upper
2
2,000 lines per
minute, single-
spaced
IBM 3800
PRINTER
Multiple
1
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
READER
N/A
1
B2
1200 cpm
IBM 3525
PUNCH
N/A
1
P3
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
2.2.2.1 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
2-7
-------
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.
2.2.2.2 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.
2.2.2.3 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.
2.3 TELECOMMUNICATIONS NETWORK
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.
2-8
-------
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
mainframe.
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
network.
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
available.
2-9
-------
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
controllers.
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
2-10
-------
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
parameter.
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
disk.
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
JUSD.BITNET.DATA(LINKS).
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 . M SERVICES TO USERS
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.
2-12
-------
NCC-IBM USER'S GUIDE
JAN 1988
Facility
NCC Computer
TSO Services
NCC-IBM User Support
Telecommunications
CICS
ADABAS
I/O Control (RTF)
Monday-Friday
24 hours
24 hours
8 AM-7 PM
24 hours
24 hours
24 hours
24 hours
Saturday
24 hours
24 hours
closed
24 hours
24 hours
24 hours
7 AM-6 PM
Sunday
until 8 PM
until 8 PM
closed
until 8 PM
until 8 PM
until 8 PM
closed
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
JUSD.HOURS.
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
JUSD.HOURS and JUSD.WICHOURS.
2.4.2
User Information Services
2.4.2.1
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
locations.
2-13
-------
NCC-IBM USER'S GUIDE JAN 1988
2.4.2.2 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.
2.4.2.3 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.
2.4.2.4 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.
2.4.2.5 On-Line Usage Aids
A group of partitioned data sets has been established to provide
comprehensive descriptions of the various software packages
2-14
-------
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
7.0.
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
available.
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.
2-15
-------
NCC-IBM USER'S GUIDE JAN 1988
2.4.3.2 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
2-16
-------
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
disapproved.
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
2-17
-------
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
restartable..
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.
2-18
-------
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
NCC-IBM.
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
requirements.
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.
2-19
-------
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
contacted.
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
2.4.3.4 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).
2-20
-------
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
JTNG.SYLLABUS.
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
duties.
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
3.3.2.
b. Prepare, log, and ship output.
2-21
-------
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
service.
2-22
-------
NCC-IBM USER'S GUIDE JAN 1983
Current micrographic rates are listed in Section 2.4.5.2
following.
A copy of the current NCC-IBM Micrographics Users Manual is in
Appendix F.
2.4.1.5 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 2.4.5.2 following.
2.4.4.6 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.
2-23
-------
NCC-IBM USER'S GUIDE
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.
2-24
-------
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$
where=
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 = $1.50 per 1000 (includes WIC)
Cards read = no charge
Cards punched = no charge
Interactive (ISO and CICS) job cost algorithm is as follows:
2-25
-------
NCC-IBM USER'S GUIDE JAN 1988
Interactive Job Cost = (CPUS + 1/0$) * 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.
2-26
-------
NCC-IBM USER'S GUIDE
2. M.S. 2 Charges for Computer-Related Services
Charges for computer-related services are listed below.
JAN 1988
COMPONENT
FY87 CHARGES
CONNECT TIME
DATA STORAGE
Public On-Line Disk Storage
Private On-Line Disk Storage
$9 . 00/Connect Hour
$.015/Track per Week
$1700/Pack per Month
STANDARD TAPE STORAGE (TMS)
NCC-IBM Tape Rental/Storage
Tape Mounts
Work Tapes (creation and
(expiration are same day)
$10.00/Tape per Month
No Charge
No Charge
FOREIGN TAPE STORAGE (non-TMS)
Days 1-5
Over 5 Days
ARCHIVE TAPE STORAGE
$2 . 00/Day
*5.00/Day
$5.00/Tape per Month
TAPE PURCHASES
$15.00/Tape
PRINTING
Lines printed locally (RTF £ WIC)
$1 .50/1000 lines
2-27
-------
KCC-IBM USER'S GUIDE
COMPONENT
CARD READER/PUNCH
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.)
Originals
Duplicates
$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
MISCELLANEOUS
Special Delivery
Data Set
Dearchiving Assistance
Cost + 10% Service Fee
$13.00/Data Set
KEYPUNCH
(Suject to increase on 7/1/88.)
Cost + 10% Service Fee
Turnaround
12 hours
24 hours
3 days
7 days
14 days
Cost per Thousand
(does not include 10% handling)
$168.00
158.00
148.00
137.00
128.00
CALCOMP PLOTTER
COURIER SERVICE
PRODUCTION CONTROL SERVICES
DASD FOR LMF SITES
$50/Plotter Hour
$260/Stop per Month
$50/Staff Hour
$1800/Site per Month
2-28
-------
NCC-IBM USER'S GUIDE JAN 1988
3.0 MANAGEMENT OF NCC-IBM SYSTEM RESOURCES
3.1 REGISTRATION PROCEDURES
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
purposes.
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.
3-1
-------
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
Initials".
3-2
-------
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
User-ID.
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
N258.
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).
3.1.3.1 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.
3-3
-------
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 3.1.3.3) 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.
3.1.3.2 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.
3.1.3.3 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.
3-M
-------
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).
3.2 SYSTEM SECURITY
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
provided.
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.
3-5
-------
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:
3-6
-------
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.
3-8
-------
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 RESOURCE MANAGEMENT
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.
3.3.1.1 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
NCC.
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.
3-9
-------
NCC-IBM USER'S GUIDE JAN 1988
3.3.1.2 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:
iiiaaaa.
or
aaaa.
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).
3.3.1.3 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)...
DATA(NAME(iiiaaaa.VSAM.xxxx.yyyy.DATA)...
INDEX(NAME(iiiaaaa.VSAM.xxxx.yyyy.INDEX))
3-10
-------
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
Administrator.
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.,
SYS2.SAS.
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
3.3.1.5 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:
iiiaaaa.SOURCE.SPROJECT
and the other
iiiaaaa.SOURCE
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
number).
3-12
-------
NCC-IBM USER'S GUIDE JAN 1988
3.3.2 Job Scheduling
3.3.2.1 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:
3-13
-------
NCC-IBM USER'S GUIDE JAN 1988
col 1 4
/*$command operands
The "/*" is omitted if the command is entered through the
console.
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
exceeded.
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.
3-14
-------
NCC-IBM USER'S GUIDE
JAN 1988
Priority
5
2 (default)
Class I
Class II
Class III
Class IV
1 (all
classes)
Weekends
Service Time
ASAP
Better than 3
Better than 2
90% of all jobs
done in < 10
minutes
90% of all jobs
done in < 30
minutes
90% of all jobs
done in < 60
minutes
90% of all jobs
done in < M hours
overnight
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
accounts.
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
S-15
-------
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
resource.
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
3-16
-------
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.
3.3.2.2 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
/*ROUTE PRINT HOLD
/*AFTER 0600
//*
3-17
-------
NCC-IBM USER'S GUIDE JAN 1988
//* CNTL(STARTAM)
//*
//START1 EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(STARTPM)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
//START2*EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB1)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
//STARTS EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB3)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BIKSIZE=80,RECFM=F,BUFNO=1)
//START4 EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBM)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BIKSIZE=80,RECFM=F,BUFNO=1)
//STARTS EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBS)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
//START6 EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB6)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
//START7 EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOB?)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
//STARTS EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1 DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBS)
//SYSUT2 DD SYSOUT=(A,INTRDR),
// DCB=(BIKSIZE=80,RECFM=F,BUFNO=1)
//START9 EXEC PGM=IEBGENER
3-18
-------
NCC-IBM USER'S GUIDE
JAN 1988
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT1
//SYSUT2
DD DISP=SHR,DSN=uuuaaaa.CNTL(JOBLAST)
DD SYSOUT=(A,INTRDR),
DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
Example of Late Night Job
//uuuPMI
//
/*ROUTE
/*AFTER
//*
//*
//*
//START!
//SYSIN
//SYSPRINT
//SYSUT1
//SYSUT2
//
//START2
//SYSIN
//SYSPRINT
//SYSUT1
//SYSUT2
//
//STARTS
//SYSIN
//SYSPRINT
//SYSUT1
//SYSUT2
JOB (aaaafimas,Bnnn),'user-name',TIME=(,3) ,
PRTY=M,MSGCLASS=P
PRINT HOLD
2300
CNTL(STARTPM)
EXEC PGM=IEBGENER
DD DUMMY
DD DUMMY
DD DISP=SHR,DSN=uuuaaaa.CNTL(STARTAM)
DD SYSOUT=(A,INTRDR),
DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
EXEC PGM=IEBGENER
DD DUMMY
DD DUMMY
DD DISP=SHR,DSN=uuuaaaa.CNTL(AMJOB2)
DD SYSOUT=(A,INTRDR),
DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
EXEC PGM=IEBGENER
DD DUMMY
DD DUMMY
DD DISP=SHR,DSN=uuuaaaa.CNTLCAMJOBZ)
DD SYSOUT=(A,INTRDR),
DCB=(BLKSIZE=80,RECFM=F,BUFNO=1)
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
3-19
-------
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
PRINT HOLD.
The last clean up job could then issue a status message to the
user informing him of any jobs that terminated in an unusual
manner.
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.
3-20
-------
NCC-IBM USER'S GUIDE JAN 1988
3.3.3.1 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
nonuse.
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
ISO:
$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.
3-21
-------
NCC-IBM USER'S GUIDE JAN 1988
3.3.3.2 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
referenced.
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
3-22
-------
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
3.3.3.3 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.
3-23
-------
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 3.3.3.2 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 3.3.3.5 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.
3.3.3.4 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
3-2M
-------
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
recommended.
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
active.
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,
3-25
-------
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.
3.3.3.5 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
3-26
-------
NCC-IBM USER'S GUIDE JAN 1988
Support.
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.
3.3.3.6 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.
3.3.3.7 Scratch
Certain types of data sets are routinely scratched from the user
disk packs. The following types of data sets are scratched
daily:
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.
3-27
-------
NCC-IBM USER'S GUIDE JAN 1988
3.3.3.8 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.
3.3.3.9 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.
3.3.3.10 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
3-28
-------
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.
3-29
-------
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.
3.3.3.11 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.
3.3.3.12 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=
3-30
-------
NCC-IBM USER'S GUIDE JAN 1988
o Name and User-ID of volume owner and an alternate
owner.
o Account code to be charged.
o Address and telephone number of owner and alternate
owner.
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
allocated.
o At least 60% of the data sets must be accessed
monthly.
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
packs:
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
3-31
-------
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.
3.3.3.13 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.
3-32
-------
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.
3-33
-------
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
information.
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;
CYL
TRK
SPACE=(blksize(primary,secondary,directory))
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:
//STEP EXEC PGM=IEBGENER
//SYSPRINT DD SYSOUT=A
//SYSIN DD DUMMY
//SYSUT1 DD DSNAME=XXXYYYY.MAY87.DAY18.TESTS.DATA,DISP=SHR
//SYSUT2 DD DSNAME=XXXYYYY.MAY87.DATACD18T5),DISP=OLD
3-34
-------
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
3-35
-------
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
follows:
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 4.1.3.3).
3-36
-------
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 3.3.1.5.
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 3.3.4.2 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;
3-37
-------
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
date.
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.
3-38
-------
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.
3.3.4.2 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
intervention.
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
3-39
-------
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
TMS.
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,
3-MO
-------
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:
3-41
-------
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:
VOL=vvvvvv[-vvvvvv][,format]
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:
DSN=data-set-name[,format]
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
generation.
Where format specifies the print format (see a. above).
Multiple control statements are allowed per execution
3-U2
-------
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.
VOL=000010-000012,SHORT
VOL=000015,LONG
DSN=iiiaaaa.z
3.3.4.3 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.
3-M3
-------
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
3-4M
-------
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),
// DISP=OLD,UNIT=TAPE,DCB=OPTCD=B
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.
3-45
-------
NCC-IBM USER'S GUIDE JAN 1988
3.3.4.4 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
number.
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
transaction.
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
shown.
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
3-46
-------
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
conversations.
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 3.3.3.1 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
3-47
-------
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).
3-M8
-------
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 TSO AND CICS ACCESS
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
as IBMEDIT or IED.)
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.
3-49
-------
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.
3-50
-------
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.)
PLEASE TYPE YOUR TERMINAL IDENTIFIERt
-NNNN-PPP-
PLEASE LOG IN = epaKCR)
P $#tt
IBMd) IS ON LINE
tso(CR)
enter logon
logon iii/kkkkkkkk acct(aaaaxxxxmuuu) (CR)
ICH7000 iii LAST ACCESS AT HH'MM:SS ON DAY-OF-WEEK,
MONTH DAY, YEAR
III LOGON IN PROGRESS AT HH:MM:SS ON MONTH DAY, YEAR
MM/DD/YY: SYSTEM BROADCAST MESSAGES SEE NEWS ALERT*
READY
Where
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
identifier.
# - 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*
3.4.1.2 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 $$LOGON.
3-52
-------
NCC-IBM USER'S GUIDE
JAN 1988
c. If $*LOGON is present, the CLIST is executed after your
logon values (User-ID, password, account, logon proc,
and logon procedure) are validated and before control is
returned to you.
A sample *$LOGON CLIST is as follows =
PROC 0 DEBUG
/*
/* LOGON PROC
/*
/* CONTROL END(ENDO)
CONTROL NOMSG NOLIST NOFLUSH
IF SDEBUG NE= THEN CONTROL MSG LIST NOFLUSH CONLIST SYMLIST
FREE F(ISPCTLO,ISPCTL1,ISPCTL2)
FDSN SYSPROC,PROC,NOW
FDSN SYSHELP,HELP,NOW
FDSN ISPPLIB,PLIB,NOW
FDSN ISPTLIB,TLIB,NOW
FDSN ISPTABL,TABL,NOW
FDSN IS?MLIB,MLIB,NOW
FDSN ISPSLIB,SLIB,NOW
FDSN ISPLLIB,LLIB,NOW
FREE FCSYSPROC,ISPPLIB,ISPTLIB,ISPMLIB, +
SYSHELP,ISPSLIB,ISPTABL,ISPLLIB)
ALLOC FCSYSPROC) DSC +
£PROC +
'JUSD.CLIST1 +
) SHR
IF SLASTCC NE= 0 THEN ALLOC FCSYSPROC) DSCfiPROC) SHR
ALLOC F(SYSHELP) DSC +
£HELP +
) SHR
IF £LASTCC NE= 0 THEN ALLOC FCSYSHELP) DSC£HELP) SHR
ALLOC FICISPPLIB) DSC +
SPLIB +
) SHR
IF £LASTCC NE= 0 THEN ALLOC FCISPPLIB) DSCfiPLIB) SHR
ALLOC FCISPMLIB) DSC +
EMLIB +
) SHR
IF fiLASTCC NE= 0 THEN ALLOC FCISPMLI.IO DSCSMLIB) SHR
ALLOC FCISPSLIB) DSC +
fiSLIB +
) SHR
IF £LASTCC NE= 0 THEN ALLOC FCISPSLIB) DSCSSLIB) SHR
/* ' x/
/* NOTE: TLIB IS THE INPUT TABLE LIBRARY. TABL IS THE */
/* OUTPUT TABLE DATSET. ONE IS NOT PERMITTED TO WRITE TO A */
/* CONCATENATED OBJECT. */
/* */
3-53
-------
NCC-IBM USER'S GUIDE JAN 1988
ALLOC F(ISPTLIB) DSC +
£TLIB +
) SHR
IF SLASTCC NE= 0 THEN ALLOC F(ISPTLIB) DSCSTLIB) SHR
ALLOC FCISPTABL) DSC +
£TABL +
) SHR
IF fiLASTCC NE= 0 THEN ALLOC FCISPTABL) DSCSTABL) SHR
ALLOC FCISPLLIB) DSC +
SLLIB +
) SHR
IF SLASTCC NE= 0 THEN ALLOC FCISPLLIB) DSCSLLIB) SHR
/* ALLOC FCISPLLIB) DSC +
/* LOAD +
/* £LLIB +
/* ) SHR
/* IF SLASTCC NE= 0 THEN ALLOC FCISPLLIB) DSCSLLIB) SHR
IF £DEBUG NE= THEN LISTA STATUS
TIME
WRITENR ENTER ISPF OPTION ===>
READ £ISPF
ISPF £ISPF
EXIT
The new user should not attempt to change this sample *$LOGON
without assistance. An improperly structured $$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:
ALLOC DACCLISLT) LIKE C'SYS2.EPA.CLIST')
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
$*LOGON.
3-SM
-------
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
satisfactory.
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).
3-55
-------
NCC-IBH USER'S GUIDE JAN 1988
3.5 RJE ACCESS TO NCC-IBM
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).
3.6 JOB ENTRY SUBSYSTEM (JES2)
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
storage
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.
3-56
-------
NCC-IBM USER'S GUIDE JAN 1988
4 . 0 PROCESSING JOBS THROUGH THE NCC-IBM SYSTEM
M.1 JOB CONTROL LANGUAGE (JCL)
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.
M-1
-------
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
User-ID.
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
«4-2
-------
NCC-IBM USER'S GUIDE
JAN 1988
by a user to further subdivide or classify computer
usage. If specified, these characters may be
alphanumeric.
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.
etime
lines
cards
forms
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.
copies
Optional positional parameter. Indicates the number
of printed output originals desired. The default is
1 .
log
linect
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.
4-3
-------
NCC-IBM USER'S GUIDE
JAN 1988
programmer
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 =
password
MSGCLASS=s
MSGLEVEL=(a,b)
TI11E=(m,s)
PRTY=p
REGION=xK
COND=(c,op)
TYPRUN=tttt
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.
14-14
-------
NCC-IBM USER'S GUIDE
JAN 1988
NOTIFY=iii
RD =
Used to request a message be sent to a user's
TSO terminal when a batch job has completed
processing.
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 =
procname,
or
//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
4.1.2.1 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
M-6
-------
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).
4.1.2.2 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
4-7
-------
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 $) characters, and must not begin
with a numeric character). The DD name must start
in column 3. No DD name can be used if
concatenation to the preceding liii statement is
desired.
c. At least one blank follows the DD name (or "//"' if
a DD name was not coded).
d. "DD".
e. At least one blank follows DD.
f. An operand (positional parameter or keyword
parameter).
g. All other required operands (keyword parameters),
separated with single commas.
h. At least one blank.
i. Any desired comments.
4-8
-------
NCC-IBM USER'S GUIDE JAN 1988
j. To continue a DD statement onto the next line, the
user may end any operand with a comma in or before
Column 71, start the next line with a // and at
least one blank, and continue listing necessary
operands. The first of these must start on or
before Column 16.
In addition to the DD name coding rules given in b. above,
several DD names have special meanings. These DD names and their
meanings are discussed in detail in the IBM MVS Extended
Architecture JCL Manual (Order No. GC28-1352) in the section
entitled DD Statement.
There are two categories of operands which may be specified on DD
statements: positional parameters, which must precede all other
operands, and keyword parameters, which may be coded in any
order. The only positional parameters are '*', 'DATA1, 'DUMMY1,
or 'DYNAM*. Lists of both types of DD statement parameters, and
a discussion of their use, may be found in the IBM MVS Extended
Architecture JCL Manual (Order No. GC28-1352).
When using card data as input which has // in columns 1 and 2,
the DD card must be as follows:
//CARD DD DATA
If the card data has /* in columns 1 and 2, a delimiter parameter
should be coded as follows:
//CARDS DD DATA,DLM=*$
(card data)
4.1.3.1 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:
4-9
-------
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),
// DCB=(RECFM=VBA,LRECL=137,BLKSIZE=6160)
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
PASS
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.
M-10
-------
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 2.2.1.6 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 4.1.3.2. 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 4.1.3.2).
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
4-12
-------
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.
4.1.3.2 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
format:
CYL index
SPACE=( TRK ,(primary,secondary,directory))
blocksize
Where:
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
4-13
-------
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
advantages:
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
cases:
o No secondary was specified when _ne data set was
created.
o The original secondary allocation is to be
overridden.
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
details).
4-14
-------
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
computation.
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.
4-15
-------
NCC-IBM USER'S GUIDE JAN 1988
CYL ind
SPACE=( TRK,(pri,sec,dir),RLSE,CONTIG,ROUND)
blksz
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.
M-16
-------
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
records.
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:
IF KEYLEN=0 THEN OVERHEAD:=15
ELSE OVERHEAD--=22
IF KEYLEN=0 THEN KEYSPAC£:=0
ELSE KEYSPACE:=(KEYLEN+12)/32
DATASPACE:=(BLKSIZE+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
storage.
4-17
-------
NCC-IBM USER'S GUIDE JAN 1988
DATASPACE is the number of 32-byte increments required for data
storage.
The number of blocks that can be stored in one track can be
computed with the following formula:
NUMBLKS:=1499/(OVERHEAD+KEYSPACE+DATASPACE)
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
records.
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.
4-18
-------
NCC-IBM USER'S GUIDE
JAN 1988
MAXIMUM BYTES
Blks/Trk
*1
2
3
14
5
6
7
8
9
10
1 1
12
13
14
15
20
25
30
W/0 Keys
N/A
23,476
15,476
1 1 ,476
9,076
7,476
6,356
5,492
4,820
4,276
3,860
3,476
3, 188
2, 932
2,676
1 ,876
1 , 396
1 ,076
PER BLOCK
W/Keys*
N/A
23, 240
15,240
1 1 ,240
8,840
7,240
6, 12C
5,256
4,584
4, 040
3,624
3,240
2,952
2,696
2,440
1 ,640
1,160
840
*The specifications for 1 block per track
are omitted due
which prohibit
32,760 bytes.
1
to software
a blocksize
limitations
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;
Wiley-Interscience
John Wiley £ Sons
605 Third Avenue
New York, NY 10016
4.1.3.3
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.
4-19
-------
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 ),
// UNIT=(TAPES,,DEFER),DCB=(RECFM=FB,LRECL=80,BLKSIZE=8000,DEN=2),
// 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=
4-20
-------
NCC-IBM USER'S GUIDE JAN 1988
DISP =
NEW
OLD
(SHR
MOD
KEEP
DELETE
, CATLG
UNCATLG
PASS
KEEP
DELETE
CATLG)
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).
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
mounted.
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.
4-21
-------
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
DCB=DEN=2.
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
exist:
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.
4-22
-------
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:
//ddname DD DSN=NONCAT.TAPE.DATA.SET,DISP=OLD,VOL=SER=021111,
// UNIT=TAPE,LABEL=(2,SL)
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
M-23
-------
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
BPI
UNIT
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.
The
4-24
-------
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
*TAPEMARK*
Second file - first data set
*TAPEMARK*
Third file - first standard trailer
*TAPEMARK*
Fourth file - second standard header
*TAPEMARK*
Fifth file - second data set
*TAPEMARK*
Sixth file - second standard trailer
*TAPEMARK*
*TAPEMARK*
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
M-25
-------
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 3.3.4.2. 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).
4.1.3.4 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
processing.
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)
4-26
-------
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
/*ROUTE PRINT HOLD
//STEP1 EXEC PGM=GKM01
//STEPLIB DD DISP=SHR,DSN=filename.loadlib
//FT06F001 DD SYSOUT=*
//*
//* MVS/EXTENDED ARCHITECTURE JCL REFERENCE, PAGE 10-98. DD
//* LABEL. SEE THE FOLLOWING DISCUSSION IN THE MANUAL.
//* THE IN SUBPARAMETER INDICATES THAT A DATA SET IS TO BE
//* OPENED FOR INPUT ONLY.
//*
//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 :
4-27
-------
NCC-IBM USER'S GUIDE JAN 1988
//OUT1 OUTPUT DEST=STLNODE.WMSMITH
//OUT2 OUTPUT CONTROL=DOUBLE
//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
comments.
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.
4.1.4.1 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.
CHARS=xxxx
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
CHARS=(xxxx,xxxx....).
CKPTLINE=nnnnn
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.
4-28
-------
NCC-IBM USER'S GUIDE JAN 1988
CKPTPAGE=nnnnn
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.
CLASS=n
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.
CONTROL=[PROGRAM/SINGLE/DOUBLE/TRIPLE]
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=nnn
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.
or
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.
4-29
-------
NCC-IBJ1 USER'S GUIDE JAN 1988
DEFAULT=Y(ES) or N(0)
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
is DEFAULT=NO.
DEST=nnnnnn
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.
FCB=xxxx
An alphanumeric value indicating the data set forms control or
carriage specifications (1-M characters).
FLASH=overlay-name
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.
or
FLASH=(overlay-name[,count])
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.
or
FLASH=(,count)
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.
FORMS=forms-name
Identifies the forms on which the SYSOUT data set is to be
M-30
-------
NCC-IBM USER'S GUIDE JAN 1988
printed or punched. form-name is 1-8 alphanumeric or national
characters.
or
FORMS=STD
Indicates a standard form.
GROUPID=output-group
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
installation-defined.
INDEX-nn
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.
JESDS=ALL
Processes the all of the jobs system-managed data sets.
JESDS=JCL
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.
or
JESDS=LOG
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.
or
JESDS=MSG
M-31
-------
NCC-IBM USER'S GUIDE JAN 1988
Processes any system messages for this job.
LINDEX=nn
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.
LIKECT=nnn
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
parameter.
MODIFY=module-name
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.
UCS=character-set-code
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
information.
WRITER=name
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
4-32
-------
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 =
CLASS=A,FORMS=6111,DEST=R3
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
4.1.14.3 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.
4.2 PROCEDURES
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 =
4-33
-------
NCC-IBM USER'S GUIDE JAN 1988
a. PROCLIB+ - private procedure libraries (see Section
4.2.8).
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.
4-34
-------
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
4-35
-------
NCC-IBM USER'S GUIDE JAN 1988
value:
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
statements.
Coding the following on the EXEC or PROC statement (see b. above)
assigns a value to a symbolic parameter:
Parameter-name=value
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:
parameter-name=
No value follows the equal sign.
The following sample illustrates an in-stream procedure set up by
M-36
-------
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 ...
//PRINT PROC LIB=,OUT=A
//GENER EXEC PGM=IEBGENER
//SYSIN DD DUMMY
//SYSPRINT DD DUMMY
//SYSUT2 DD SYSOUT=£OUT
//SYSUT1 DD DISP=SHR,DSN=IJKAAAA..SLIB
// PEND
// EXEC PRINT,LIB='FILE(MEMB)'
M . 2 . 5 Overriding Catalogued Procedures
4.2.5.1 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 =
parameter.procstepname=value
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 =
FARM.STEP1=CBA
To nullify this same FARM, he may code the following:
PARM.STEP1=
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:
U-37
-------
NCC-IBM USER'S GUIDE JAN 1988
// EXEC PROCA,REGION.STEP1=150K,
// 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
nullified.
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.
4.2.5.2 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 4.2.5.1.)
4-38
-------
NCC-IBM USER'S GUIDE JAN 1988
14.2.5.3 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.
M-39
-------
NCC-IBM USER'S GUIDE JAN 1988
COBUCL: Procedure to compile and link a COBOL source
//DEFAULTS PROC PRINT=A,PUNCH=B, SYSPRINT/SYSPUNCH
(1) // CREGION=320K,CPARM=, COMPILER REGION/FARM DEFAULTS
// LREGION=256K,LPARM='XREF,LIST,LET1,SIZE1=,SIZEZ=
//* LINKAGE DEFAULTS
//COB EXEC PGM=IKFCBLOO,REGION=£CREGION,
// PARM ='LIB,XREF,SOURGE,NOZWB,£CPARM1
(2) //SYSLIN DD DSN=£OBJMOD,DISP=(MOD,PASS),UNIT=SYSDA,
// SPACE=(3120,(100,10))
//SYSPRINT DD SYSOUT=£PRINT
//SYSPUNCH DD SYSOUT=£PUNCH
//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) )
//*
(3) //LKED EXEC PGM=HEWL,PARM='£LPARM,SIZE=(£SIZE1,£SIZE2) ' ,
(4) // REGION=£LREGION,COND=(4,LT,COB)
//SYSLIB DD DSN=SYS1.COBLIB,DISP=SHR
(2) //SYSLIN DD DSN=£OBJMOD,DISP=(OLD,DELETE)
// DD DDNAME=SYSIN
(5) //SYSLMOD DD DSN=£LODMOD(MAIN),DISP=(,PASS),UNIT=SYSDA,
// SPACE=(CYL,(2,2,1))
//SYSPRINT DD SYSOUT=£PRINT
//SYSUT1 DD UNIT=(SYSDA,SEP=(SYSLIN,SYSLMOD)),
// 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.
M-40
-------
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...
(1) // EXEC COBUCL,COND.LKED='(4,LE,COB)'
(2) //COB.SYSIN DD DSN=IIIAAAA.ASM(COBPROG),DISP=SHR
(3) //LKED.SYSLMOD DD DSN=IIIAAAA.LOAD,DISP=SHR,UNIT=,SPACE=
(4) //LKED.SYSIN DD *
(5) ALIAS COBPROG
(6) NAME PROGRAM
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.
4-41
-------
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
SYS2.USERPROC
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
follows:
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.
4-42
-------
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:
(TSO/2ED)
EDIT procedure-name CNTL NEW
*** build JCL in workspace
SAVE 'SYS2.PROCADD(procedure-name)'
(TSO/ISPF)
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
SYS1.USERPROC.
(Batch)
//useridXX JOB ...
//STEPA EXEC PGM=IEBUPDTE,PARM=NEW
//SYSPRINT DD SYSOUT=A
//SYSUT2 DD DSN=SYS2.PROCADD,DISP=SHR
//SYSIN DD DATA
./ ADD NAME=procedure-name
//* iii/aaaa J.SMITH (202)999-9999
//procedure-name PROC
Place Balance of JCL Procedure Here (Remove all /*
statements)
./ ENDUP
/*
A word of caution: Each new procedure must be uniquely named.
4-M3
-------
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'
RETPROC
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
number).
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
follows
//ALLOC EXEC PGM=IEFBR14
//PROCALOC DD UNIT=DISK,DISP=(NEW,CATLG,DELETE),
// 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
given.
4-M4
-------
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. . .
//PROCLIB DD DISP=SHR,DSNAME=iiiaaaa.PROCLIB1
// DD DISP=SHR,DSNAME=aaaa.PROCLIB2
//stepname EXEC procname
M.3 OUTPUT PROCESSING
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
4-M5
-------
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
parameters.
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
4-46
-------
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").
4.3.2.1 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).
4-47
-------
NCC-IBM USER'S GUIDE
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.
Form
8381
C181
C151
Lines/Page
Char/Line
61
-------
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.
4.3.2.3 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
example:
//iii JOB (acctuid,room),'prog name',MSGCLASS=F
//stepname EXEC . . .
//PRINTOUT DD SYSOUT=F
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 =
/*JOBPARM FORMS=8381
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
printers.
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.
4-49
-------
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.
4-50
-------
NCC-IBM USER'S GUIDE
JAN 1988
4.3.3.1
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.
BURST
CHARS
COPIES
DEST=N3
FCB
FLASH
FORMS
MODIFY
DCB=OPTCD=J
LINECT
/*JOBPARM
x
x
X
X
X
/XOUTPUT
x
X
X
X
X
X
X
X
SYSOUT DD
x
X
X
X
X
X
X
X
X
Notes:
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.
4.3.3.2
Forms for the WIC Laser Printers
The forms and print densities available at WIC are as follows:
Form
631 1
6351
6371
C15
C17
838
839
C18
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"
Default
FCB
6
6
6
C151
C151
8381
8381
8381
Default
Char/in.
10
10
10
10
10
15
15
15
Default
Lines/in.
6
6
6
12
12
8
8
12
Default
Lines/pg
61
61
61
61
61
61
61
61
4-51
-------
NCC-IBM USER'S GUIDE
The Forms Control Buffer is explained below;
JAN 1988
FCB =
6
8
C151
UNI6
C181
8381
Lines/in,
6
8
8
12
12
8
12
Max Lines/pg
for 57 lines
for 4 more
80
120
60
90
for 58 lines
for 3 more
Form
Length
1 1"
1 1"
11"
1 1"
8.5"
8.5"
Valid Forms
6311, 6351,
6371 , C151 ,
C171
631 1
6371
C17 1
631 1
637 1
C17 1
6351 ,
C151 ,
6351 ,
C151 ,
6311, 6351,
6371 , C1S1,
C171
8381, 8391,
C181
8381, 8391,
C181
Note that there are three forms with 3-hole punch: 6371, C171,
and 8391.
U.3.3. 3
Fonts
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).
M.3.M
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
unit.
High-quality output is produced on a variety of sizes and weights
of cut-sheet paper or forms appropriate to the immediate needs.
M-52
-------
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
immediately.
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
(G544-3089-0)
Planning for the Location of the IBM 3820 Page Printer
(G544-3090-1)
Planning for Advanced Function Printing Software for MVS
(GS44-3091-0)
4-53
-------
NCC-IBM USER'S GUIDE JAN 1988
Ordering Supplies for the IBM 3820 Page Printer
(G544-3092-1)
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
(S544-3084-0)
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 )
4.3.5
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.
4.3.6
Output Distribution an<
loom Parameters
Output routed to the central site (NCC) will be processed by
Operations Support and distributed by Data Processing Support
4-54
-------
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
sooner.
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.
4-55
-------
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
output.
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 4.3.6.1). 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
delay.
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 .
14-56
-------
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
paper.
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
M-57
-------
NCC-IBM USER'S GUIDE
*
-------
NCC-IBM USER'S GUIDE
JAN 1988
TN: All the characters in the PN train plus the
following special characters =
Character
Hexadecimal
Description
*
t
abcdef ghi
jklmnopqr
stuvwxyz
L
cent sign
exclamation point
lower case letters
left brace
less or equal
superscript left paren
superscript plus
dagger
lower case letters
right brace
lozenge
superscript right paren
plus/minus
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
dash
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.
0123M56789
j
X
X
X
X
X
X
'81
X
X
X
X
X
1 91
X
X
X
X
X
X
A2
X
X
X
X
X
'BO
X
X
X
X
X
'MA'
'5A'
1 - X'89'
'8B1
'8C'
'8D'
'8E'
'8F1
i _ x'99'
'9B'
'9C'
'9D'
'9E1
'9Ff
'A1 '
1 - X1 A9'
' AB'
'AC'
'AD'
'AE'
'AF1
1 - X'B9'
'BB'
'BC'
'BD'
'BE'
'BF'
4-59
-------
NCC-IBM USER'S GUIDE
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
1
3
PN
TN
Third digit: Specifies the paper desired as follows:
Value Paper Description
1
2
5
7
8
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.
4-60
-------
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
4-61
-------
NCC-IBM USER'S GUIDE
JAN 1988
Form
No. A B
2301
2302
2303
2304
2305
15/1
-1/2 1-7/1
15/1
15/1
-1/2 1-7/1
2306 1-1/2 1-7/1
Dimensions
C
6
6
6
6
6
6
3-1/2
3
3-1/2
3-1/2
it
4
D E
1
1-1/2 3. 1
1 1/8
1 1/8
1-1/2 4-1/8
1-1/2
Description
of Labels
1 -up
3-up
3-up
4-up
3-up
1 -up
Figure 4-5. Self-Stick Labels
Note that self-stick labels cannot be used with the IBM 3800
printer.
4.3.9
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).
4.3.10
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 2.2.1.4.
4-62
-------
NCC-IBM USER'S GUIDE JAN 1988
5. 0 SUPPORTED UTILITY FUNCTIONS
5.1 ISO UTILITIES
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)
5-1
-------
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)
[COPIES(nnn)]
OUTLM(outlimit)
HOLD/NOHOLD
LIST/NOLIST
PRINT/NOPRINT [NOMSG]
CAPS/ASIS [UCS(image) HTEXT 1
FORMS(form)
HEAPING/NOMEADING
[SNUM]
[ASA ]
[VOLUME (volser ) ]
BURST
Required - dslist, bin
Alias - PRINTO, LISTOFF, LISTO
Defaults - CLASS(A), COPIESM). NOHOLD, LIST, PRINT,
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
5-2
-------
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
printed.
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
information.
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.
5-3
-------
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'.)
5-U
-------
NCC-IBM USER'S GUIDE
JAN 1988
System Displays
COMMAND
DA
DT
DS
DC
[B/S/T ]
LOcate (level)
Input Queue
Display
Output fiueue
Display
D2
DI
AI
HI
DF
[class Hdest ]
[class Hdest ]
[class Hdest ]
DO [class Hdest ]
AO [class Hdest ]
HO [class ] [dest]
DESCRIPTION
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
class.
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
[purge]
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
jobname
NEWCLASS(class)'.
5-5
-------
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.
5-6
-------
NCC-IBM USER'S GUIDE
JAN 1988
TYPE
Miscellaneous
Job-Related
List-Related
COMMAND
H/HElp cmd/msgid
E/End
STatus jobid
JCL jobid [R]
DESCRIPTION
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' .
Exit.
Display job.
List job JCL. The R
option specifies
that 'raw'
(unconverted) JCL be
displayed (input JCL
as it was
submitted).
List JES2 job log.
Only available after
job has completed
execution.
List job messages.
JES2 DD summary
(ddname/dsid).
List JES2 data set.
Obtain dsid values
by using the DD
subcommand.
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
data.
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
M-PF19
T/TOp
B/BOttom
Move forward # lines.
Move backward * lines.
* is 1 .
Max to top of data set
Top of current data
set.
Bottom of current data
5-7
-------
NCC-IBM USER'S GUIDE
JAN 1988
Logging
M-PF20
RLn
SAve dsname
set.
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
set.
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
specified.
EXAMPLE:
If job XXX1 is
subcommand may
JES2 job number
be either:
136, then the DD
WARNING:
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).
5-8
-------
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
in JUSD.NODE.LOCATION.DATA.
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
5-9
-------
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:
5-10
-------
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
JUSD.JSTS.GUIDE).
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.
5.1.7.1 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.
5.1.7.2 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
//SYSPRINT DD SYSOUT=*
//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)
DISPCCATALOG,DELETE) RECFM(FB) LRECLC80) BLKSIZE(4240)
SPACE(1,1) CYL DIRC20) DSORG(PO) PARMS(MSG=A)
VOLUMECvolser )
/EOT
5.1.7.3 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
executes.
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.
5-12
-------
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
exist.
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
"uid".
Commands to Check Progress. BDT provides a variety of commands
that allow querying of data transfer transactions.
5.1.7.4 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
(GC23-0223-0)
MVS/Bulk Data Transfer, Version 2, File to File Transaction
Guide (SC23-0225-0)
5-13
-------
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
(SC23-0227-0)
5.2 IBM UTILITIES
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 .
5-14
-------
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
volumes.
o Renames data sets or members of data sets on DASD
volumes.
o Catalogs or uncatalogs data sets.
5-15
-------
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:
// EXEC PGM=IEHPROGM
//SYSPRINT DD SYSOUT=class
//ddname DD VOL=REF=SYS1.SVCLIB,DISP=OLD
//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.
DSNAME=name
,VOL=device=list
[ , PURGE]
[,MEMBER=name]
b. RENAME changes the name or alias of a data set or
member residing on a direct access volume.
DSNAME=name
,VOL=device=list
,NEWNAME=name
[,MEMBER=name]
c. CATLG generates an entry in the index of a catalog.
DSNAME=name
,VOL=device=list
d. UNCATLG removes an entry from the lowest level
index of the catalog.
DSNAME=name
5-16
-------
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
present.
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
5-17
-------
NCC-IBM USER'S GUIDE JAN 1988
specific User-ID/account combination, the following JCL may be
used :
// . . . JOB. . .
//STEP1 EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=A
//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
JCL:
5-18
-------
NCC-IBM USER'S GUIDE JAN 1988
//. . .JOB. . .
//BUILDGDG EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=A
//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.
5-19
-------
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:
// EXEC PGM=IEBGENER
//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)
//SYSUT2 DD DSN=iiiaaaa.SAVEDECK,DISP=(NEW,CATLG),
// 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
functions.
The basic JCL format is as follows:
5-20
-------
NCC-IBM USER'S GUIDE JAN 1988
// EXEC PGM=IEBCOPY
//SYSPRINT DD SYSOUT=class
//SYSUT3 DD UNIT=SYSDR,SPACE=(TRK,(1,1))
//SYSUT4 DD UNIT=SYSDR,SPACE=(TRK,(1,1))
//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;
COPY OUTDD=SYSUT2,INDD=SYSUT1
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.
OUTDD=ddname
,INDD=ddname1[,ddname2,...]
or
ddname1[ ,ddname2 ] [ ,(ddname3,R). . . ]
or
((ddname1,R)[ ,ddname2 ] . . . )
[,LIST=NO]
5-21
-------
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, . . . )
or
((name 1,neunarae,R),...)
or
((name 1,neuname),...)
or
((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
replaced.
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
5-22
-------
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
//STEPCOPY EXEC PGM=IEBCOPY
//SYSPRINT DD SYSOUT=A
//INPDS DD DISP=SHR,DSNAME=iiiaaaa.oldfile
//OUTPDS DD DISP=(NEW,CATLG,DELETE),
// DSNAME=iiiaaaa.newfile
// SPACE=(6160,(30,6,5)),UNIT=DISK
//SYSUT3 DD UNIT=SYSDR,SPACE=(TRK,(1,1))
//SYSUT4 DD UNIT=SYSDR,SPACE=(TRK,(1,1))
//SYSIN DD *
COPY OUTDD=OUTPDS,INDD=INPDS
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
preallocated.
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 =
// EXEC PGM=IEHMOVE
//SYSPRINT DD SYSOUT=class
//SYSUT1 DD UNIT=SYSDR,DISP=OLD,SPACE=(TRK,(1,1))
//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.
5-23
-------
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.
DSNAME=name
,TO=device=list
[ , FROM = device = list]
[,UNCATLG]
[ ,CATLG ]
[ ,RENAME = name ]
t,FROMDD = ddname ]
[ ,TODD = ddname ]
[ ,UNLOAD ]
b. COPY (or MOVE) PDS copies (or moves) a partitioned
data set.
PDS=name
,TO=device=serial
[ ,FROM = device = serial ]
[ ,EXPAND = nn ]
[ ,UHCATLG ]
[ ,CATLG ]
[ ,RENAME = new-name ]
[ ,FROMDD = ddname 1
[ ,TODD = ddname ]
[ ,UNLOAD ]
5-24
-------
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
set.
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
obtained.
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
[,MEMBER=membname]
[,FROM=device=list]
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
operation.
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.
5-25
-------
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
copied.
REPLACE DSNAME=dsname
[,MEMBER=member]
t,FROM=device=volume]
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
//SYSPRINT DD SYSOUT=class
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:
5-26
-------
NCC-IBM USER'S GUIDE
Parameter SYSUT1
COPY ISAM
UNLOAD ISAM
LOAD
PRINTL
PRINTL,N
SE2
ISAM
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 =
//STEP1 EXEC PGM=IEBISAM,[PARM=UNLOAD]
//SYSPRINT DD SYSOUT=A
//SYSUT1 DD DSN=XXXAAAA.IS,DISP=SHR
//SYSUT2 DD UNIT=TAPE,DISP=(,PASS),DSN=£ISTAPE,
// DCB=(RECFM=FB,LRECL=80,BLKSIZE=8000)
//STEP2 EXEC PGM=IEBISAM,PARM=LOAD
//SYSPRINT DD SYSOUT=A
//SYSUT1 DD DSN=£ISTAPE,DISP=SHR
//SYSUT2 DD DSN=XXXAAAA.IS,DISP=OLD,DCB=DSORG=IS
5.2.8
IEHLIST
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 =
// EXEC PGM=IEHLIST
//SYSPRINT DD SYSOUT=A
//ddname DD VOL=SER=dsvol,UNIT=DISK,DISP=OLD
//SYSIN DD *
(control statements)
5-27
-------
NCC-IBM USER'S GUIDE JAN 1988
The control statement is as follows =
DUMP
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
parentheses.
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
modules.
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
5-28
-------
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
5.2.9.1 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
££OBJMOD:
//SYSLIN DD DSN=££OBJMOD,DISP=(OLD,DELETE)
// DD *
ENTRY CSECT1
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:
5-29
-------
NCC-IBH USER'S GUIDE JAN 1988
//SYSLIN DD DSN=££OBOMOD,DISP=(OLD,DELETE)
// 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.
5.2.9.2 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.
5.2.9.3 SYSUT1
SYSUT1 is a sequential work data set used by the Linkage Editor.
5.2.9.4 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=FA,LRECL=121,BLKSIZE=121)
DCB=(RECFM=FBA,LRECL=121,BLKSIZE=n*121) where n is
an integer not greater than MO.
5.2.9.5 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.
5-30
-------
NCC-IBM USER'S GUIDE JAN 1988
5.2.9.6 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
//LKED.SYSLMOD DD DSN=IIIAAAA.LOAD,DISP=SHR,UNIT=,SPACE=
//LKED.SYSLIN DD *
INCLUDE OBJECT
INCLUDE SYSLMODCLOAD1 ,LOAD2 )
ENTRY MAIN
NAME NEWMOD(R)
//LKED.OBJECT DD DSN=IIIAAAA.OBJ,DISP=SHR
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
IDCAMS).
5-31
-------
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
addressing.
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
OLD.
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 =
5-32
-------
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 OTHER SUPPORT FUNCTIONS
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
SORT
Some of the enhancements incorporated in the latest version of
SYNSCORT are:
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
5-33
-------
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
GUSD .SYNCSORT.DATA.
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-
Context.
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
tasks.
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 "...".
5-34
-------
NCC-IBM USER'S GUIDE JAN 1988
INPUT DATA:
UNI2UE NUMBER: UN-01
TITLE: HAZARDOUS WASTES AND HAZARDOUS DUMPS
KWIC INDEX:
...AND HAZARDOUS DUMPS UN-01
HAZARDOUS WASTES AND HAZARDOUS DUMPS UN-01
HAZARDOUS WASTES AND... UN-01
HAZARDOUS WASTES AND HAZARDOUS DUMPS 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.
KWOC INDEX:
DUMPS
HAZARDOUS WASTES AND HAZARDOUS DUMPS UN-01
HAZARDOUS
HAZARDOUS WASTES AND HAZARDOUS DUMPS UN-01
WASTES
HAZARDOUS WASTES AND HAZARDOUS DUMPS UN-01
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).
5-35
-------
NCC-IBM USER'S GUIDE
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
iiiaaaa.CNTLCMEMBER).
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.
5-36
-------
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
5-37
-------
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.
// EXEC TAPESCAN,VOL=vvvvvv,PARM='LIST20,NOHEX'
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
files/tapemarks.)
to dump file x, SKIPTMn, where n = (3* (. x-1 ) ) +1 , MAXTMm, m = n+1
// EXEC TAPESCAN,VOL=vvvvvv,
// PARM='LIST75,SKIPTM4,MAXIMS'
3. LIST summaries of the first 30 blocks of the first 2 files
on an NL tape.
// EXEC TAPESCAN,VOL=vvvvvv,LBLTYP=NL,
5-38
-------
NCC-IBM USER'S GUIDE JAN 1988
// PARM='LIST30,MAXTM2'
4. LIST the first 20 blocks of each file on a SPERRY tape.
// EXEC TAPESCAN,VOL=vvvvvv,LBLTYP=NL,OPTCD=g,
// 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 .
5.3.6.1 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 $$LOGON and type "S" to select.
Since each screen refresh rewrites each line of your terminal
screen, this process is slow and inefficient. For a method of
accomplishing the same logical functions with only one screen
refresh, refer to the example in the discussion of MYED below.
5-39
-------
NCC-IBM USER'S GUIDE JAN 1988
5.3.6.2 Using MYLISTC
Note: The following examples refer to CLISTS in the file
JUSD.CLIST. You can copy these into your ISO session
by copying JUSD.CLIST($LOGON) into your file
CLIST($$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.
5.3.6.3 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
activity.
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
of JUSD.CLIST(MYLISTC).
Minimizing the number of screen refreshes, at the TSO READY
prompt, type the following:
1. ISPF 6
This bypasses the ISPF menu panel.
2. MYED 'JUSD.CLIST1;S MYLISTC
5-40
-------
NCC-IBM USER'S GUIDE JAN 1988
You are editing the PDS JUSD.CIIST and the member
MYLISTC.
3. GETSET
Invoke the ISPF edit macro JUSD.CLIST(GETSET ).
Steps 2 and 3 can be combined in the following single command:
MYED 'JUSD.CLIST1;s MYULISTC;GETSET
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:
F ISPEXEC
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:
S-M1
-------
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 =
END;END
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
terminates.
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 =
5-142
-------
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
managed.
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.
5-43
-------
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»
5.3.10.1 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
developed.
5-M4
-------
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
parameters.
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
classes.
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.
5.3.10.2 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
5-45
-------
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)
CALL 'DPT1.LOADCPRG1)'
EXIT (filastcc)
S-46
-------
NCC-IBM USER'S GUIDE JAN 1988
6 . 0 PROGRAMMING LANGUAGES
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
6-1
-------
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
applications.
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
6-2
-------
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.
6.1 ASSEMBLER LANGUAGE
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:
ASMvfff
where v is the version (H) and fff is an acronym for the function
performed by the procedure. The procedures are=
6-3
-------
NCC-IBM USER'S GUIDE
Assembler HV2
ASMHC
ASMHCL
ASMHCLG
ASMHCG
JAN 1988
Function
assemble only
assemble and
link-edit
assemble, link,
execute
and
assemble and execute
without link-edit
alias for ASMGCG
link-edit and
execute
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 =
6-4
-------
NCC-IBM USER'S GUIDE
Parameter Default
PRINT A
PUNCH
DUMMY
CPARM
CREGION
LREGION
LPARM
GREGION
SIZE1
NO BATCH
250K
250K
LET,XREF,LIST
NCAL
or
LET,MAP,NOCALL
250K
(unspecified)
SIZE2
(unspecified)
SIZE
8192K
JAN 1988
Meaning
SYSOUT class for printed
output
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
assembler
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
parameter)
The number of bytes to be
reserved for buffers by the
Linkage Editor (used as the
second component of the SIZE
parameter)
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.
6-5
-------
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
(SC26-4037)
o Assember H Version 2 Application Programming
Language Reference (5026-4036)
o MVS/XA Supervisor Services and Macro Instructions
(GC28-1154)
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.
6-6
-------
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 =
6-7
-------
NCC-IBM USER'S GUIDE
Parameter
PRINT
PUNCH
CPARM
CREGION
LREGION
GREGION
LPARM
GPARM
SIZE1
SIZE2
SIZE
Default
A
B
(unspecified)
320K
256K
450K
XREF,LIST,LET
or
XREF,LIST,HAP
(unspecified)
(unspecified)
(unspecified)
8192K
OAN 1988
Meaning
SYSOUT class for printed output
SYSOUT class for object deck
FARM field for the COBOL compiler
which has a default of
'LIB,XREF,SOURCE,NOZWB,NODUMP,£CPARM'
The region available to the
compiler
The region available to the Linkage
Editor
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.
6.2.2
Publications
Further information concerning parameters which may be supplied
to the COBOL compiler or to the Linkage Editor is contained in
the publications listed below.
6-8
-------
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
COBOL II.
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
II.
o That any application procured or migrated from
outside sources be converted to COBOL II.
6-9
-------
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 =
i
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
levels.
6-10
-------
NCC-IBM USER'S GUIDE JAN 1988
Major enhancements added to the COBOL conversion aid are as
follows:
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
library.
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
-------
NCC-IBM USER'S GUIDE
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).
6.3.2
Publications
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
Editor.
VS FORTRAN Version 2
(GC26-M219)
General Information
VS FORTRAN Version 2 = Language and Library
Reference (SC26-4221)
VS FORTRAN Version 2- Programming Guide
(SC26-4222)
VS FORTRAN Version 2- Interactive Debug Guide and
Reference (SC26-M223)
VS FORTRAN Version 2: Reference Summary
CSX26-3751-2)
OS/VS Linkage Editor and Loader (GC26-3813)
MVS/XA Linkage Editor and Loader (GC26-4011)
6-12
-------
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
execute
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 =
6-13
-------
NCC-IBM USER'S GUIDE
Parameter Default
PRINT A
PUNCH DUMMY
CPARM
CREGION
LREGION
LIB
LPARM
GREGION
SIZE1
SIZE2
GPARM
SIZE
(unspecified)
256K
256K
BASE
LET,LIST,XREF
or
LET,MAP
256K
(unspecified)
(unspecified)
(unspecified)
192K
JAN 1988
Meaning
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
1 DECK,£CPARM'
The region available to the
compiler
The region available to the Linkage
Editor
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.
6-14
-------
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
(SC33-0006)
o OS PL/1 Optimizing Compiler: Messages (SC33-0027)
o OS/PL/1 Optimizing Compiler: TSO User's Guide
(SC33-0029)
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
(SC28-8308)
6-15
-------
NCC-IBM USER'S GUIDE JAN 1988
7 . 0 APPLICATION PACKAGES
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
package.
o The appropriate STEPLIB DD statements are included.
7-1
-------
NCC-IBM USER'S GUIDE JAN 1988
7.1 DATA BASE SYSTEMS
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
DBMS .
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
Environment
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).
7-2
-------
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
below.
7-3
-------
NCC-IBM USER'S GUIDE JAN 1988
7.1.2.1 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).
7.1.2.2 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.
7-4
-------
NCC-IBH USER'S GUIDE JAN 1988
Documentation on BASIS can be found in the on-line data set
1 JUSD.BASIS.DATA1 .
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 RETRIEVAL AND REPORTING SYSTEMS
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
computations
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
7-5
-------
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
file(s).
o SORT records in ascending and/or descending
sequences.
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
7-6
-------
NCC-IBM USER'S GUIDE JAN 1988
b. IRS Training Supplement: Sample Problems and
Solutions.
c. Data set JUSD.IRS.DATA has examples typical of IRS
applications.
7.2.2 EASYTRIEVE PLUS
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
tasks.
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
through EASYTRIEVE PLUS using SYNCSORT.
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.
7-7
-------
NCC-IBM USER'S GUIDE JAN 1988
7.3 MATHEMATICAL AND STATISTICAL PACKAGES
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
histograms
o Bivariate F and t tests and comparison of cells
o Scatter plots, probability plots
o Missing values estimates, patterns, and
correlations
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
options
o Nonparametric statistical analysis
o Factor analysis, canonical correlation,
discriminant analysis
o Cluster analysis on cases, variables, or blocks
7-8
-------
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
9002M.
See the on-line usage aid data set JUSD.BMDP.DATA for BMDP
documentation and sample BMDP programs. Sections 2.4.2.6
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
capabilities.
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.
7-9
-------
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:
HELP;
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
JUSD.SAS.SAMPLE.SOURCE.
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
7-10
-------
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
ESP.
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
inexpensively.
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
7-12
-------
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:
//SYSLIB DD DSN=SYS2 . PLMATH .PROCDRS,DISP=SHR
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.
7-13
-------
NCC-IBM USER'S GUIDE JAN 1988
7.4 COMPUTER GRAPHICS SOFTWARE
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
required.
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
documentation!
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.
7-14
-------
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
JUSD.TELAGRAF.DATA.
7.M.2 CUECHART
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.
7.U.3 DATA CONNECTION
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.
c. EXTERNAL PROGRAM CONNECTION - Links TELL-A-GRAF
directly with user-written applications programs.
7-15
-------
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
data set JUSD.DATACONN.DATA.
7.4.4 TELLAPLAN EXPERT
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.
7-16
-------
NCC-IBM USER'S GUIDE JAN 1988
For more information, see the on-line documentation file
JUSD.ORGCHART.DATA.
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
7-17
-------
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
language.
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
coordinates.
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.
7-18
-------
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
PREVIEW.
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
7-19
-------
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
set JUSD.HARVARD.DATA.
7-20
-------
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
Houston Instruments SYS2.COMPLOT.LOAD JUSD.COMPLOT.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
support.
More information on GDDM can be found in the data set
'JUSD.GDDM.DATA'.
7.5 FILE MANAGEMENT SYSTEMS
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.
7-21
-------
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
library
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
7-22
-------
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.
7.6 DEBUGGING AND OPTIMIZING SYSTEM - OPTIMIZER III
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
OPTIMIZER, DETECTOR, and ANALYZER.
The OPTIMIZER component produces a smaller and more efficient
program by eliminating redundant and unnecessary machine
instructions at the object code level.
7-23
-------
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
problems.
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 SIMULATION AND MODELING SYSTEMS
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:
7-24
-------
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
meanings.
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.
7-25
-------
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.
7.8 TEXT EDITORS - SCRIPT
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.
7.9 FOCUS, A FOURTH GENERATION LANGUAGE
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
reports.
ISPF users can invoke FOCUS through the primary menu or by
7-26
-------
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.
7-27
-------
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.
7-28
-------
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
encrypted.
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
7-29
-------
NCC-IBM USER'S GUIDE JAN 1988
code conversion. This compatibility will reduce
training costs and permit better use of personnel
resources.
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
auare:
a. Like all Fourth Generation Languages, FOCUS can
place huge demands on computing resources. Good
file design and control of file size are therefore
essential.
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.
7-30
-------
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
Files
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
7-31
-------
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
command:
MEGACALC
7-32
-------
NCC-IBM USER'S GUIDE JAN 1988
A MegaCalc support group has been established. Call (714)
474-0800 for more information.
7-33
-------
NCC-IBM USER'S GUIDE APPENDIX A
APPENDIX A: TELEPHONE NUMBERS
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
EPA-NCC
MD-34
Research Triangle Park, NC 27711
Data Processing Support Services
EPA-NCC
MD-24
Research Triangle Park, NC 27711
Disk Librarian
FIMAS Office
EPA-NCC
MD-34C
Research Triangle Park, NC 27711
I/O Control Desk
or
Shift Supervisor (NCC)
NCC Communications and Training
NCC-IBM Production Control
Neu Agreements
Operations and Problems
NCC-IBM System Status Recording
(919)
(FTS)
541-2693
629-2693
(919) 541-2385
(FTS) 629-2385
(800) 334-9700
(919) 541-2385
(FTS) 629-2385
(800) 334-9700
(919)
(FTS)
541-1 137
629-1 137
(919) 541-2385
(FTS) 629-2385
(800) 334-9700
(919) 541-7862
(FTS) 629-7862
(800) 334-2405
(919)
(FTS)
(919)
(FTS)
(919)
(FTS)
541-2377
629-2377
541-1494
629-1494
541-4732
629-4732
A-1
-------
NCC-IBM USER'S GUIDE
APPENDIX A
JAN 1988
NCC User Support
EPA-NCC
MD-34B
Research Triangle Park, NC 27711
NCC Network Control Facility
EPA-NCC
MD-24
Research Triangle Park, NC 27711
Network Support Group
Tape Librarian
EPA-NCC
MD-24
Research Triangle Park, NC 27711
TSSMS Office
EPA-NCC
MD-34C
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)
(FTS)
541-3182
629-3182
(919) 541-2385
(FTS) 629-2385
(800) 334-9700
(919)
(FTS)
541-3629
629-3629
(202) 488-5900
(800) 424-9067
(202) 488-5980
(202) 488-5970
NCC Key Personnel: Office Telephone Numbers
EPA NDPD
Don Fulford
Maureen Johnson
(919) 541-3970
(FTS) 629-3970
or
(919) 541-3483
(FTS) 629-3483
(919)
(FTS)
541-2501
629-2501
A-2
-------
NCC-IBM USER'S GUIDE APPENDIX B JAN 1988
APPENDIX B= SAMPLE FORMS
USER PROFILE WORKSHEET
Please enter/verify:
NAME:
ADDRESS:
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
Hardware
Telecommunications System
Technical Advisories
Software (All Software)
Systems Software
Utilities
Compilers (All Compilers)
FORTRAN
PL/1
COBOL
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
EPA
MD-34B
Research Triangle Park, NC 27711
Attn: User Profile
B-1
-------
NCC-IBI1 USER'S GUIDE APPENDIX B JAN 1988
SAMPLE FORMS
NCC-IBM USER'S GUIDE READERS' COMMENTARY
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
organization.
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
examples):
o Do you require a response?
This form, together with any supporting material, should be
forwarded to:
NCC-IBM User Support
EPA-NCC
MD-3MB
Research Triangle Park, NC 27711
Thank you for your comments.
B-2
-------
NCC-IBM USER'S GUIDE APPENDIX B JAN 1988
REQUEST FOR REFUND
PLEASE COMPLETE THE USER INFORMATION SECTION OF THIS REPORT IN
INK, ATTACH ALL PERTINENT BACKUP MATERIAL, AND SEND TO= NCC-IBM
USER SUPPORT/EPA-NCC/MD-34B/RESEARCH TRIANGLE PARK, N.C.
27711/ATTN: REFUND REQUEST
USER INFORMATION
USER NAME
USERID EPA ACCT CODE REGION NO
PHONE MAIL CODE REGIONS 1-10
HDQTRS 11
STREET LAB 12
CONTRACTOR 13
CITY, STATE, ZIP OTHER 14
REQUESTED REFUND ADJUSTMENTS
JOB START END
NAME JOB # DATE TIME TIME AMT ACCT
1 .
2 .
3.
4.
USER REASON FOR REQUESTING REFUND
1 .
2 .
3.
TYPE OF MATERIAL ATTACHED
DATE RECEIVED REFUND NO
B-3
-------
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
NDPD TELECOMMUNICATIONS SERVICE REgUEST
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
NCC/NETWORK SUPPORT GROUP
EPA
MD-24A
RTF, NC 27711
B-13
-------
NCC-IBJ1 USER'S GUIDE APPENDIX B
NDPD TELECOMMUNICATIONS SERVICE RE2UEST
JAN 1988
PAGE 1
ADP Coordinator Name
Title
Signature
Contact Site Location
Organization
Building
Address
City State
Requestor or Contact Person
Contact
Commercial Phone
EPA Project Officer
Phone
Location
Room Number
Zip Code
Contact
FTS Phone
Phone (
NARRATIVE EXPLANATION OF REQUESTED ACTION
Requested Completion Date
This change will be= Permanent
If Temporary, what is duration? Months
Temporary
From
RETURN TELECOMMUNICATIONS SERVICE REQUEST TO:
To
NCC/NETWORK SUPPORT GROUP
EPA
MD-2MA
RTP, NC 27711
i 1
I FOR ODP USE ONLY |
I TAG RECEIPT: |
I TSR No. |
I Date |
I Time |
I Title |
L 1
B-14
-------
NCC-IBM USER'S GUIDE APPENDIX B JAN 1988
INSTRUCTIONS FOR PAGE 1
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.
B-15
-------
NCC-IBM USER'S GUIDE APPENDIX B JAN 1988
FULL-SCREEN APPLICATION TERMINAL EfiUIPMENT PAGE 2
FULL-SCREEN EQUIPMENT ACTION
guantitv/Manuiacturer/Model
Terminals Printers Controllers
New
Expand
Move*
Delete
Existing Full-Screen Equipment:
Terminals Printers
Controllers
ASCII TERMINALS USING PROTOCOL CONVERTER
Gtv/Manufacturer/Mod
Terminals
New
Expand
Move*
Delete
Protocol Converter Ports
Existing Full-Screen Equipment: Terminals
*MOVE INFORMATION
From: Building To: Building
Ports
Room
Room
Attention
Telephone
Address
City
State
Zip
Attention
Telephone
Address
City
State
Zip
B-16
-------
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.
B-17
-------
NCC-IBM USER'S GUIDE APPENDIX B
GRAPHICS AND ASCII (TTY) TERMINAL EfiUIPMENT
GRAPHICS TERMINAL E6UIPMENT
gtv/Manufacturer/Model
Terminals Printers
New
Expand
Move*
Delete
JAN 1988
PAGE 3
Host Type
ASCII (TTY) TERMINAL EgUIPMENT
New
Expand
Move*
Delete
gty/Manufacturer/Model
Terminals Printers Host Type
*MOVE INFORMATION
From: Building
Room
Attention
Telephone
Address
City
State
To-- Building
Room
Zip
Attention
Telephone
Address
City
State Zip
B-18
-------
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.
B-19
-------
NCC-IBM USER'S GUIDE APPENDIX B
RJE REGISTRATION FORM
OAN 1988
PAGE 4
TERMINAL TYPE
Terminal Manufacturer
Terminal Model Number
Line Protocol (BSC/SDLC)
Line Printer Speed
Existing Line Type
(Lease/Dial)
Host Access
TERMINAL PERIPHERALS
Number of Readers
Number of Printers
Number of Punches
Protocol Equivalent
2770, 3781
?'780
3780
S360/20 Mod 4,5
S360/20 Mod 5,6
S360/22, 25, etc.
1 130
System 3
3770
3270-Attached:
3262
3287
NTR
1004
i i
PRINTER SETUP PR1 PR2 PR3
Print Positions
Lowercase
Separator PAGE
Drain Start
TERMINAL FEATURES
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
B-20
-------
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
request.
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
sign-on.
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.
B-21
-------
NCC-IBM USER'S GUIDE APPENDIX B
PROJECTED E2UIPMENT ACTIVITY
FILE TRANSFER (PERSONAL COMPUTER)
JAN 1988
PAGE 5
I Receiving I From I To I File
Application I Device I Location I Location I Size
I I I I
I I I I
I I I I
L
1
1
1 Freq.
1
1
1
REMOTE JOB PRINTING (RJE)
1
1
1
1
1
1
1
1
1
1
Application 1
1
1
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
BULK DATA TRANSFER (HOST TO HOST)
I
Application I
From I To I File I I Time
Location I Location ! Size I Freq. I of Day
I
I
I
B-22
-------
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 . )
B-23
-------
NCC-IBM USER'S GUIDE
APPENDIX B
JAN 1988
PAGE 6
TERMINAL ACTIVITY BY FUNCTION AND HOST SYSTEMS
HOST SYSTEMS
APPL. FUNCTION
DATA ENTRY
Hours/Day
No. of Terminals
Avg . Input
Char/Trans ....
Avg . Output
Char/Trans ....
Avg . No . of
Trans/Hour ....
Avg . No . of
Trans . Peak
Hour
PROGRAM DEV.
Hours/Day
No. of Terminals
Usage (L, A, H)*
GEN. TIMESHARING
Hours/Day
No. of Terminals
Usage (L, A, H)*
GRAPHICS
HoursXDay
No. of Terminals
Usage (I, A, H)*
CHAR. PNTR OTPUT
No. Char. Pntrs
Average Lines/Hr
Avg Char/Prt
Line
ISO
LOG RTP
SPF
LOG RTP
:
GIGS
LOG RTP
SPERRY
DEC
PRME
OTH
*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
B-24
-------
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
interfaces.
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
Park.
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.
B-25
-------
NCC-IBM USER'S GUIDE APPENDIX C JAN 1988
APPENDIX C- JES2 CONTROL STATEMENTS
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:
SIGNOFF
SIGNON
Statements for modifying JOB statements =
JOBPARM
NOTIFY
Statements controlling job selection:
AFTER
BEFORE
CNTL (EXCLUSIVE or SHARED)
Statements to control routing of output;
ROUTE
Statements for communication with the operator:
MESSAGE
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.
C-1
-------
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
C-2
-------
NCC-IBM USER'S GUIDE
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
completed.
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
/*CNTL
/*CNTL
resource,EXC
or
resource,SHR
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.
/*AFTER JOBNAME1
/^BEFORE JOBNAME2
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*AFTER JOBNAME
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/^BEFORE JOBNAME
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
/*CNTL RESOURCE,disp
(disp = SHR or EXC)
C-3
-------
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
JOBNAMEO
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
C-4
-------
NCC-IBM USER'S GUIDE
APPENDIX C
JAN 1988
255 lines)
LINES=nnnn
where nnnn specifies the estimated number of
lines to be printed, in thousands (from 0 to
9999)
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
parameters.
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
C-5
-------
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
delimiter.
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
C-6
-------
NCC-IBM USER'S GUIDE
APPENDIX C
JAN 1988
into separate sheets. N indicates
that the printed output is to be in
continuous, fanfold mode. N is the
default.
CHARS=xxxx
CKPTLNS=nnnnn
CKPTPGS=nnnnn
COPIES=nnn
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.
C-7
-------
NCC-IBM USER'S GUIDE
APPENDIX C
JAN 1988
COPYG=nnn
DEST=xxxx
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 =
Rnnn
RMnnn
RMTnnn
NnnRmmmm
NnnnRmmm
NnnnnRmm
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.
FCB=xxxx
FLASH=overlay-name
FLASHC=count
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.
C-8
-------
NCC-IBM USER'S GUIDE
FORMS=xxxx
INDEX=nn
LINDEX=nn
APPENDIX C
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.
MODTRC=trc
UCS=xxxx
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
manual.
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
statements.
C-9
-------
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:
/*ROUTE PRINT N7
SIGNOFF Statement: The SIGNOFF statement is used to disconnect
an RJE station from the NCC-IBM computer system. Its form is:
col 1
/*SIGNOFF
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).
C-10
-------
NCC-IBH USER'S GUIDE APPENDIX D JAN 1988
APPENDIX D: JES2 OPERATOR COMMANDS (SP2.1.5)
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
D-1
-------
NCC-IBM USER'S GUIDE
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=XEBA
*DN,e=XE2A,IBM or $DN,XE2A
Vanilla:
NCC JES2:
and
Vanilla:
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
RMTnnn
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:
Rnnn.xxi
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.
D-2
-------
NCC-IBM USER'S GUIDE APPENDIX D JAN 1988
JES2 OPERATOR COMMANDS SUMMARY
*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
device
*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
device
D-3
-------
NCC-IBM USER'S GUIDE APPENDIX D JAN 1988
COMMENTS =
o The $R command will only affect jobs awaiting
output.
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
or
$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.
D-M
-------
NCC-IBM USER'S GUIDE APPENDIX D
COMMANDS CONCERNING THE JOB BUEUE
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
Output
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
2=XEex,IBM
fi=PPU,IBM
fi=HOLD,IBM
fi=OUT,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
utilization.
Operand: e=XE2,IBM
fi=XEBx,IBM
2=PPU,IBM
2=HOLD,IBM
2=OUT,IBM
D-S
-------
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
$DF,J=Jxxxx
where xxxx is the job number.
D-6
-------
NCC-IBM USER'S GUIDE APPENDIX D
COMMANDS CONCERNING SPECIFIC JOBS
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
Command:
Function =
Examples
*DJ joblist
$D 'jobname'
Display identifying and status information for
the specified job(s).
$DJ1-9999
$DJ18,19,22-25
$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
D-7
-------
NCC-IBM USER'S GUIDE
APPENDIX D
JAN 1988
Examples
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 .
*AJ5-8
$A 'XYZ051
Cancel Job
Command:
Function:
Examples
$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.
$CJ5-8
$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
D-8
-------
NCC-IBM USER'S GUIDE APPENDIX D
COMMANDS CONCERNING A DEVICE
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
D-9
-------
NCC-IBM USER'S GUIDE
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.
Example:
Cancel Device
Command:
Function
R178.PR1
$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
D-10
-------
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.
Examples
Drain Device
Command:
Function:
$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
-------
NCC-IBM USER'S GUIDE APPENDIX D
MISCELLANEOUS JES2 COMMANDS
JAN 1988
Send a Message to the Computer Operator
Command: $DMRO,'message text1
Function;
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.
Examples
*DMRO
$DMRO
Change Job Routing
Command =
'Please re-queue job 1234 to R1231
'I canf't seem to make it work'
*RPRT
or
*RPUN,J=jobid,D=RMTnnn
or
*RALL
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
D-12
-------
NCC-IBM USER'S GUIDE APPENDIX D JAN 1988
computer site for output processing.
Examples: $RALL,J=J1117,D=R121
*RALL,
-------
NCC-IBM USER'S GUIDE APPENDIX D JAN 1988
JOB
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:
WS=(W,R,2,PMD,LIM,F,T,C/P)
D-14
-------
NCC-IBM USER'S GUIDE APPENDIX E JAN 1988
APPENDIX E= TELECOMMUNICATIONS INFORMATION
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).
TROUBLESHOOTING AND REPORTING PROBLEMS
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
problem
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.
E-1
-------
NCC-IBM USER'S GUIDE APPENDIX E JAN 1988
FAST BUSY SIGNAL
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.
SLOW BUSY SIGNAL
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.
RING-NO-ANSWER
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.
TONE-NO-RESPONSE
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.
RANDOM "GARBAGE" CHARACTERS
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
E-2
-------
NCC-IBM USER'S GUIDE APPENDIX E JAN 1988
identifier.
DROPS (DISCONNECT/LOSS OF CARRIER)
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
extension.
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
telephone>)
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.)
E-3
-------
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
JUSD.SYSTEM.MESSAGES(TELCOM).
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.
E-M
-------
NCC-IBM USER'S GUIDE APPENDIX F JAN 1988
APPENDIX F: NCC-IBM MICROGRAPHICS
Introduction
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
film.
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
F-1
-------
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:
F-2
-------
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.
//STEP1 EXEC PGM=IEFBR14
//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 =
//STEP1 EXEC PGM=IEFBR14
//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
processing.
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.
F-3
-------
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
center.
5. The contracted micrographics center processes the
application=
F-M
-------
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
assistance.
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:
F-5
-------
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
application.
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
film
F-6
-------
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
appropriate.
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
Titling
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 characterthe
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
table:
24X80 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
F-7
-------
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.
Indexing
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
F-8
-------
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
alternative.
F-9
-------
WIC REQUEST FOR MICROGRAPHICS
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
DATE:
NAME OF APPLICATION:
INDIVIDUAL USER: NAME
ADDRESS
ZIP CODE _
PHONE NO. ROOM NO.
USER I.D. ACCOUNT NO
TYPE OF APPLICATION
.NEW APPLICATION (TEST STATUS)
CHANGE APPLICATION
CURRENT STATUS: TEST
PRODUCTION
SERVICE REQUESTED
.ORIGINAL MICROFICHE
.ORIGINAL ROLL FILM
.SOURCE DOCUMENT MICROFILMING
.DUPLICATION OF MICROFICHE
DUPLICATION OF ROLL FILM
SPECIAL INSTRUCTIONS
MICROGRAPHICS Page 1 of 5
-------
WIC MICROFORMS INPUT-OUTPUT SPECIFICATIONS
INPUT (Required for microfiche or roll film request)
MAGNETIC TAPE CHARACTERISTICS:
IBM OR OTHER (SPECIFY)
LABELED OR UNLABELED (L or U)
TRACK (7 or 9)
BPI (556,800,1600 ONLY)
PARITY (ODD OR EVEN-RESUIRED FOR 7-TRACK TAPES ONLY)
RECORD LENGTH (MAX 256)
BLOCK SIZE (MAX 9000 CHARACTERS)
CHARACTER SET (EBCDIC, ASCII, OR OTHER)
NUMBER OF PRINT POSITIONS
CARRIAGE CONTROL POSITION (if exist, usually pos. 1)
POSITION NUMBER OF FIRST PRINT CHARACTER (if
carriage control exist, usually pos. 2)
CARRIAGE CONTROL TYPE (ASA, MACHINE, OTHER)
LINES PER FRAME/PAGE (INCLUD. PAGE HEADERS)(MAX 86)
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.)
CHANNEL
LINE NUMBER
OUTPUT
MICROFORMS REBUIREMENTS '
NUMBER OF COPIES RE2UESTED
MICROFICHE OR ROLL FILM
IMAGE (COMIC OR CINE) (NORMALLY COMIC FOR FICHE)
FRAMES/ROW (HORIZONTAL)
FRAMES/COLUMN (VERTICAL)
REDUCTION FACTOR (2 MX,42X,48X)
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
L
n
I6I7I8I9IAIBICI
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
-------
WIC MICROFORMS INPUT-OUTPUT SPECIFICATIONS
OUTPUT(continued from page 2)
MICROFICHE TITLE FORMAT:
SPECIFY HOW TITLE SHOULD APPEAR CIRCLE AND NUMBER EACH EXTRACT
I !
I I
I I
_______________ ___ _ J
TITLE EXTRACT FIELDS DATA
FIELD 123456
p- _________.«» ^m -|
LINE NUMBER
BEGINNING PRINT
POSITION
FIELD LENGTH
EXAMPLE FROM YOUR
DATA
HOLD/NORMAL
PAGE HEADER FORMAT'
ELEMENT
BEGIN
PRNT POS
HEADER ELEMENT DESCRIPTION
LINE 1
Skip lines
after linel
(BLANK LINES)
.
LINE 2
Skip lines
after Iine2
(BLANK LINES)
LINE 3
Skip lines
after lineS
(BLANK LINES)
LINE 4
(Skip lines
after Iine4
1 (BLANK LINES)
L
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
MICROGRAPHICS
Page 3 of 5
-------
WIC MICROFORMS INPUT-OUTPUT SPECIFICATIONS
OUTPUTC continued from page 3)
INDEXING REQUIREMENTS:
SUBKEY
1
i
SUBKEY
2
SUBKEY
3
SUBKEY
ll
' I
BEG PRINT POSITION FOR KEY
KEY LENGTH IN CHARACTERS
LINE TO BEGIN KEY SEARCH
LINE TO END KEY SEARCH
KEY MASK CHARACTER
PRINT LINE POSITION FOR
MASK
MASK LENGTH IN CHARACTERS
LITERAL STRING (4 CHAR MAX)
MASK CHARACTERS: Numeric = 9; Alphabetic = A; Alphanumeric =X,
Special Character = S; Literal = L.
SOURCE DOCUMENT MICROFILMING;
TITLE OF DOCUMENT
NUMBER OF PAGES
PAGE DIMENSIONS
NUMBER OF MICROFILM COPIES DESIRED
SPECIFY FILM CARTRIDGE TITLE, IF ANY
FORMAT REDUCTION RATIO
COMIC 24X
CINE 25X
27X
CARTRIDGE TYPE
3M
KODAK
REEL
MICROGRAPHICS
Page 4 of 5
-------
WIC MICROFORMS INPUT-OUTPUT SPECIFICATIONS
FICHE BREAK REQUIREMENTS:
COND 1 COND 2 COND 3
COND 4
BREAK ON MATCH OR
CHANGE?
START NEW FICHE WITH
CURRENT OR NEXT
FRAME
LINE TO BEGIN SEARCH
LINE TO STOP SEARCH
BEGINNING PRINT POS.
OF BREAK FIELD
LENGTH IN CHARACTERS
OF BREAK FIELD
BREAK EVERY OCCURRENCE
OF MATCH?
MATCH FIELD 1
MATCH FIELD 2
MATCH FIELD 3
MATCH FIELD M
MATCH FIELD 5
MATCH FIELD 6
MICROGRAPHICS
Page 5 of 5
-------
WIG COM SIGN-OFF SHEET
CONTROL NO. DATE
NAME OF APPLICATION
USER: NAME
ADDRESS
ZIP CODE
PHONE NO. ROOM NO.
EXPECTED FIRST PRODUCTION DATE
EXPECTED FRESUENCY/VOLUME
APPROXIMATE TIME OF WEEK, MONTH, OR YEAR NORMALLY RE2UIRED
TEST SATISFACTORY
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
category.
PLEASE RETURN COM SIGN-OFF SHEET TO
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
APPENDIX G: NCC SOFTWARE SUMMARY
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
APPENDIX H= INSTRUCTIONS FOR PRINTING
SE2UENTIAL DATA SETS OR MEMBERS OF PARTITIONED DATA SETS
For Printing in upper case only to SYSOUT=A=
BATCH:
//...JOB...
/*JOBPARM LINES=111
// EXEC PRINTSE2,LIB='data-set-name1,DEST = RMTnnn[ ,COPIES
ISO-'
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)]
where
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
-------
NCC-IBM USER'S GUIDE APPENDIX I
APPENDIX I-' ROUTE CODES
JAN 1988
Note: The following codes are used in the Route Code column to
indicate destination:
D
F
M
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.
ROUTE
CODE
BIN
NUMBER
DISTRIBUTION
SITE
B
nnn
800
801
802
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
22102
803
Crystal Mall, EPA, 1921 Jefferson Davis
Highway, Building Two, Room 1121, Arlington, VA
22212
804
805
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
806
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
22212
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
20460.
Page 1-2
-------
NCC-IBM USER'S GUIDE
APPENDIX I
JAN 1988
823
82(4
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
initials.
M
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
APPENDIX J= GRAPHICS EXAMPLES (OFF-LINE)
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
APPENDIX K: EPA/NCC TRAINING COURSE SYLLABUS
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 ***
TITLE: INTRODUCTION TO ADP
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
presented.
LENGTH: 1 day
PRERE2UISITES: None
Page K-1
-------
NCC-IBM USER'S GUIDE APPENDIX K JAN 1988
TITLE: FUNDAMENTALS OF INSTRUCTION
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
PRERE2UISITES: None
*** NCC MAINFRAMES ***
TITLE: INTRODUCTION TO IBM FULL-SCREEN PROCESSING
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
PRERE2UISITES: None
TITLE: ISPF FOR IBM USERS
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
TITLE: INTRODUCTION TO IBM JOB CONTROL LANGUAGE
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
explained.
LENGTH: 2 days
PRERE2UISITES: INTRODUCTION TO IBM FULL-SCREEN PROCESSING or
equivalent experience.
TITLE: TSO FOR FULL-SCREEN USERS
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
PREREfiUISITES: INTRODUCTION TO IBM FULL-SCREEN PROCESSING or
equivalent experience.
TITLE: ISO FOR NON FULL-SCREEN USERS
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
TITLE: INTRODUCTION TO TSO/ISPF FOR PROGRAMMERS
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
discussed.
LENGTH: 3 days
PREREBUISITES= Proficiency using a non-IBM computer system.
TITLE: WRITING COMMAND PROCEDURES (CLIST's)
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
PREREBUISITES: INTRODUCTION TO IBM FULL-SCREEN PROCESSING and
TSO FOR NON FULL-SCREEN USERS
TITLE: ADVANCED IBM FULL-SCREEN PROCESSING
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
PREREBUISITES: WRITING COMMAND PROCEDURES or TSO/ISPF FOR
PROGRAMMERS
TITLE: ISPF DIALOG MANAGER
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
PREREUISITES: WRITING COMMAND PROCEDURES or INTRODUCTION TO THE
LOGICAL MAINFRAME
Page K-4
-------
NCC-IBM USER'S GUIDE APPENDIX K JAN 1988
TITLE: INTRODUCTION TO THE LOGICAL MAINFRAME
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
PRERE2UISITES: None
TITLE: INTRODUCTION TO RACF
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.
TITLE: TELL-A-GRAF
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
users.
LENGTH: i day
PRERE2UISITES: None
Page K-5
-------
NCC-IBM USER'S GUIDE APPENDIX K JAN 1988
TITLE: FOCUS REPORTING
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.
TITLE: FOCUS APPLICATION DEVELOPMENT
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
TITLE: INTRODUCTION TO WORD PROCESSING
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
TITLE: ADVANCED WORD PROCESSING
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
PREREfiUISITES: INTRODUCTION TO WORD PROCESSING or equivalent
experience .
TITLE: INTRODUCTION FOR MANAGERS
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
PREREfiUISITES: None
TITLE: INTRODUCTION TO IBM PC
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
PREREQUISITES: None
TITLE: INTRODUCTION TO LOTUS 1-2-3
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
PRERE2UISITES: None
Page K-7
-------
NCC-IBH USER'S GUIDE APPENDIX K JAN 1988
TITLE: ADVANCED LOTUS 1-2-3
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.
TITLE: INTRODUCTION TO dBASE III
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
PREREQUISITES: None
TITLE: ADVANCED dBASE III
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
PREREfiUISITES: INTRODUCTION TO dBASE III or equivalent working
knowledge and experience with dBASE III.
TITLE: INTRODUCTION TO dBASE III PLUS
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
PREREfiUISITES: INTRODUCTION TO THE IBM PC or equivalent
experience.
Page K-8
-------
NCC-IBJ1 USER'S GUIDE APPENDIX K JAN 1988
TITLE: WORDSTAR
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
PREREQUISITES: None
TITLE: INTRODUCTION TO WORDPERFECT FOR BEGINNING PC USERS
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
covered.
LENGTH: " 2 days
PREBUISITES: None
TITLE: INTRODUCTION TO WORDPERFECT FOR EXPERIENCED PC USERS
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
THE IBM PC
TITLE: ADVANCED WORDPERFECT
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
PRERE2UISITES: Working knowledge of WORDPERFECT OR INTRODUCTION
TO WORDPERFECT
Page K-9
-------
NCC-IBM USER'S GUIDE APPENDIX K JAN 1988
TITLE: PC EXECUTIVE SEMINAR
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
PREREfiUISITES: None
PRIME
TITLE: PRIME ORIENTATION
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
PREREfiUISITES: None
TITLE: TEXT
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
PRERE2UISITES: PRIME ORIENTATION
Page K-10
-------
NCC-IBM USER'S GUIDE APPENDIX K JAN 1988
TITLE: INTRODUCTION TO INFO
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
PREREBUISITES-- PRIME ORIENTATION
TITLE: ADVANCED INFO
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
files.
LENGTH: 3 days
PRERESUISITES: INFO
TITLE: INFO-VERSAGRAPH
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
PRERE2UISITES: PRIME ORIENTATION and INFO
Page K-11
-------
NCC-IBM USER'S GUIDE APPENDIX K JAN 1988
TITLE: INFO-TEXT
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
PRERE2UISITES: PRIME ORIENTATION and INFO
TITLE: INTRODUCTION TO 20/20
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
PREREQUISITES: PRIME ORIENTATION
TITLE: WORDMARC
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
PREREQUISITES: PRIME ORIENTATION
Page K-12
-------
NCC-IBM USER'S GUIDE APPENDIX K OAN 1988
TITLE: VAX COURSE 1
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
PREREfiUISITES: None
TITLE: VAX COURSE 2
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
PRERE2UISITES: VAX Course 1
TITLE: VAX/FOCUS
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
-------
NCC-IBM USER'S GUIDE APPENDIX K
COMPUTER-BASED TRAINING COURSES
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.
COURSE DESCRIPTION
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
APPENDIX L= INFORMATIONAL DATA SETS
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 -
JUSD.ADABAS.DATA
JUSD.ADABAS.FORKS.TEXT
JUSD.ADRLIB.DATA
JUSD.ASSEMH2.DATA
- B -
JUSD.BADPROCS
JUSD.BASIC.DATA
JUSD.BASIS.DATA
JUSD.BASISK2.JCL
JUSD.BASISK2.PROCLIB
JUSD.BDT.DATA
JUSD.BILLING.DATA
JUSD.BITNET.DATA
JUSD.BMDP.DATA
JUSD.BMDP.MANUALS?
JUSD.BMDP.V1987.DATA
- C -
JUSD.CALCOMP.DATA
JUSD.CALCOMP.TESTDATA
JUSD.CAM.DATA
JUSD.CANCEL.LOG
JUSD.CDBA.FORMS
JUSD.CDBMS.FORMS
JUSD.CDF.DATA
JUSD.CHAR3800.DATA
JUSD.CHAR3800.DOC
JUSD.CICS.DATA
JUSD.CLIST
JUSD.CLIST2
JUSD.CNTL
JUSD.COBAID.DATA
JUSD.COBOLAID.DATA
JUSD.COMPILE.JCL
JUSD.COMPLOT.DATA
JUSD.CONVERT.GUIDE
JUSD.CONVERT.HELP
JUSD.CUECHART.DATA
JUSD.C2NC.DATA
Page L-1
-------
NCC-IBM USER'S GUIDE APPENDIX L JAN 1988
- D -
JUSD.DATACONN.DATA
JUSD.DASD.DATA
JUSD.DBASE.DATA
JUSD.DC2.DATA
JUSD.DFEF.DATA
JUSD.DIALOG.CNTL
JUSD.DIALOG.LOAD
JUSD.DIALOG.OBJ
JUSD.DISSPLA.DATA
JUSD.DS
JUSD.DSNLISTN
- E -
JUSD.EASYT.DATA
JUSD.EASYTREV.PLUS.DATA
JUSD.ESP.DATA
JUSD.EXECS.CLIST
- F -
JUSD.FLOBOL.DATA
JUSD.FOCMSTER.DATA
JUSD.FOCUS.DATA
- G -
JUSD.GDDM.DATA
JUSD.GEOMAP.DATA
JUSD.GLIM.DATA
JUSD.GLIM.IN
JUSD.GLIM.LOAD
JUSD.GRAPH.DMP29.DATA
JUSD.GRAPH.SASGOUT.DEMOS
JUSD.GRAPH.SAS5GOUT.DEMOS
JUSD.GRAPH.TABLES
JUSD.GRAPH.TAGDM1.META
JUSD.GRAPH.TAGDM2.META
JUSD.GRAPH.TAGDM3.META
JUSD.GRAPH.TAGDMM.META
JUSD.GRAPH.TEKM106.DATA
JUSD.GRAPH.TEK4170.DATA
JUSD.GRAPH.TEK4170.KERMIT
- H -
JUSD.HARVARD.DATA
JUSD.HARVARD.TESTDATA
JUSD.HELPWC.S2K
JUSD.HOURS
Page L-2
-------
NCC-IBM USER'S GUIDE APPENDIX L JAN 1988
- I -
JUSD.IBM.DOCUMENT.DATA
JUSD.IBM.EDITOR.DATA
JUSD.IBM3800.DATA
JUSD.IBM3800.TESTDATA
JUSD.IBM3820.DATA
JUSD.IGL.DATA
JUSD.IMSL.DATA
JUSD.INDEXA
JUSD.IRS.DATA
JUSD.ISPF.DATA
JUSD.ISPF.GUIDE
JUSD.ISPF.LIB.NAMES
JUSD.ISSCOGKS.DATA
JUSD.IXR.DATA
- J -
JUSD.JCL.CONVERT
JUSD.JCL.MANUAL
JUSD.JES.UPDATE.DATA
JUSD.JES328X
JUSD.JOBSCHED
JUSD.TSTS.GUIDE
- K -
JUSD.KERMIT.DATA
JUSD.KWIDOC
- L -
JUSD.LINMOD.DATA
JUSD.LINMOD.V1985.DATA
JUSD.LMF.DATA
JUSD.LMF.GUIDE
JUSD.LMF.UTILITY
- M -
JUSD.MAG.GUIDE
JUSD.MARKIV.DATA
JUSD.MEGACALC.DATA
JUSD.MEGACALC.VM1 .DATA
JUSD.MVS.PROBLEMS
JUSD.MVSXA.PROBLEMS
Page L-3
-------
NCC-IBM USER'S GUIDE APPENDIX L JAN 1988
- N -
JUSD.NATCON.DATA
JUSD.NATURAL.DATA
JUSD.NATURAL.EXAMPLE
JUSD.NEWLOGON.DATA
JUSD.NEW2UEUE.DATA
JUSD.NEWS
JUSD.NOARCH.DATASETS
JUSD.NODE.LOCATION.DATA
JUSD.NONIUS
JUSD.NOT.STANDARD
JUSD.NOTUSE90.DATASETS
- 0 -
JUSD.OLD.ALERTS.DATA
JUSD.ONLINE.DOC
JUSD.OPENLINE.DATA
JUSD.OPTIII.DATA
JUSD.OPTMRS.DATA
JUSD.ORGCHART.DATA
- P -
JUSD.PANVALET.DATA
JUSD.PC.PUBLIC.LIB
JUSD.PC70.DATA
JUSD.PHONES
JUSD.PHONES.DATA
JUSD.PLIX.DATA
JUSD.POLICIES.MANUAL
JUSD.PREDICT.DATA
- 2 -
JUSD.2ED.DATA
JUSD.2ED.DOC
- R -
JUSD.RACF
JUSD.RACF.DATA
JUSD.RACF.ORIG
JUSD.REFUND
JUSD.REFUNDS.DATA
JUSD.RMTLIST
JUSD.RTISAS.DATA
Page L-M
-------
NCC-IBM USER'S GUIDE APPENDIX L JAN 1988
- S -
JUSD.SAS.DATA
JUSD.SAS.GEOMAP.DATA
JUSD.SAS.OLD.SAMPLE
JUSD.SAS.RICDEMO
JUSD.SAS.SAMPLE.SOURCE
JUSD.SAS.USAGE
JUSD.SASGRAPH.SAMPLE.SOURGE
JUSD.SAS5.DATA
JUSD.SCRIPT.DATA
JUSD.SDSF.DATA
JUSD.SOFTWARE.DATA
JUSD.SOFTWARE.INDEX
JUSD.SPACE1
JUSD.SPERRY.MANUAL
JUSD.SPSS.DATA
JUSD.STC8380.DATA
JUSD.SUPER.NATURAL.DATA
JUSD.SURFACE2.DATA
JUSD.SYNCSORT.DATA
JUSD.SYSABEND
JUSD.SYSTEM.MESSAGES
JUSD.SYS3.USERPROC
JUSD.S2K.DATA
JUSD.S2K.DOCUMENT
- T -
JUSD.TAPESCAN
JUSD.TAPESCAN.DATA
JUSD.TEKTRONX.DATA
JUSD.TELAGRAF.DATA
JUSD . TELAGRAF.TAGLRN
JUSD.TELAPLAN.EXPERT.DATA
JUSD.TELAPLAN.RESOURCE.DATA
JUSD.TRIP.DATA
JUSD.ISO
JUSD.TSOCLIST
JUSD.TSOE.DATA
JUSD.TSOLINE.GUIDE
- U -
JUSD.UNIRAS.UNIEDIT.DATA
JUSD.UNIRAS.UNIMAP.DATA
JUSD.USA.MAPS.DATA
JUSD.USERS.GUIDE
JUSD.UTILITY.DATA
- V -
JUSD.VSAM.DATA
JUSD.VSCOBOL2.DATA
JUSD.VSFORT2.DATA
Page L-5
-------
NCC-IBM USER'S GUIDE APPENDIX L JAN 1988
- W -
JUSD.MIC.SOFTWARE
JUSD.WORDSCAN.DATA
- Z -
JUSD.ZETA.DATA
Page L-6
-------
NCC-IBM USER'S GUIDE GLOSSARY JAN 1988
GLOSSARY
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 itemThat 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
request.
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.
Glos-1
-------
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
functions.
ALC: Assembler Language Coding, the symbolic coding language
used for machine-level programming of IBM 360 and 370 series
computers.
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
volume.
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.
Glos-2
-------
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
operator.
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.
conversational).
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
Glos-3
-------
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
perforations.
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.
Glos-M
-------
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
characteristics.
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.
Glos-5
-------
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
operands.
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).
Glos-6
-------
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.
i
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
them.
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
9.2.
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.
Glos-7
-------
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.
Glos-8
-------
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)
Glos-9
-------
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
applications.
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
Glos-10
-------
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
forth.
generation data set: One of the data sets in a generation data
group.
group: A collection of RACF users who can share access
authorities for protected resources (equivalent to user
accounts).
group authority: An authority that relates to a type of function
a user can perform in a group. The group authorities are
USE, CREATE, CONNECT, and JOIN.
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
qualfier.
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
operator.
instruction: The smallest unit of work which a computer may
perform.
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
Glos-12
-------
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
transfer.
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.
Glos-13
-------
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
(q.v.).
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
minute.
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
Glos-14
-------
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
characters.
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.
Glos-15
-------
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
Glos-17
-------
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
record.
plotter: A device which draws lines on paper under program
control.
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
Glos-18
-------
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
profile.
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
results.
real storage management-1 Routines that control the allocation of
pages in real storage. Abbreviated RSM.
Glos-19
-------
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
Glos-20
-------
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
error.
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
above.
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
device.
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.
Glos-21
-------
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
output.
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
storage.
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
intercommunication.
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.
Glos-22
-------
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.
Glos-23
-------
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.
Glos-24
-------
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
ADABAS 7-2
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
Archive
data set 3-21
Assembler
available at NCC 5-3
B
Backup
procedures 3-8
BASIC
interpreter 6-15
publications 6-15
BASIS 7-4
Batch mode
password 3-7
Batch processing 5-44
Battelle's Automated Search Information System...see BASIS
Index-1
-------
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
BLKSIZE
on DD statement 4-13, 4-24
space allocation 3-32, 4-17
BMDP see Biomedical
Computer
Programs
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
Automatic
Mapping
Cancelled jobs 2-17
Card parameter
on JOB statement 4-3
on JOBPARM statement C-4
Cards
costs 2-28
Cartographic Automatic Mapping 7-20
Catalog
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
EASYTRIEVE PLUS 7-7
ESP 7-11
FORTRAN 6-11
GPSS 7-25
IMSL 7-12
library 4-33
naming 4-43
Index-2
-------
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
Charges
for services 2-24
Charges-see costs 2-24
CICS 3-55
performance goals 3-55
service level objectives 3-55
Class
of SYSOUT data sets 4-46
Classes 2-20
Cleaning tape 2-6
CLIST 3-52, 5-44
CLISTS 5-39
CNTL control statement
EXCLUSIVE C-2
SHARED C-2
COBOL 6-6
catalogued procedures 6-7
publications 6-8
COM F-1
Commerce , Department of 3-2
Compilation
available compilers 6-1
COBOL 6-6
FORTRAN 6-11
PL/1 6-13
Compress 3-28
Computer
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
Index-3
-------
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
Configuration
mainframe 2-2
Connect time
costs 2-27
CONTIG
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
Conventions
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
Courses
on-line syllabus 2-20
training 2-20
Customer Information Control System see CICS
Index-4
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
D
Daily statistics 5-44
DASD
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 CONNECTION 7-15
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
DBACCESS 7-4
DCB
on DD statement 4-12, 4-23
DC2 7-23
DD Name
on DD statement 4-8, 4-10, 4-20
Index-5
-------
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
DEN
on DD statement 4-24
Density
tape data sets 4-24
Department of Commerce 3-2
Device-Driven Subroutine Libraries 7-21
Devices
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
Disk
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
DISP
on DD statement 4-10, 4-20
DISSPLA 7-17
Documentation
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 CONNECTION 7-16
data set 3-30
DC2 7-23
dearchive 3-24
EASYTRIEVE PLUS 7-8
ESP 7-11
Index-6
-------
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
SYNCSORT 5-3M
SYSTLM 2000 7-4
TELLAPLAN EXPERT 7-16
TSO/E 3-49
usage aids 2-14
VSAM 5-32
DPSS 2-21
Drum
devices 2-3
DSSUBMIT 5-36
Catalogued procedures 5-36
E
EASYTRIEVE PLUS 7-7
Econometrics, software
ESP 7-10
Education Services 2-20
Electronic spreadsheet 7-32
Enhancements
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
Index-7
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
Execute statement 4-5
EXPDT
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
Files
Input Only 4-26
Read Only 4-26
FIHAS 3-5
FOCUS 7-26
interactive 5-45
Foreign tape
costs 2-27
Foreign Tape Reports 3-46
Foreign Tapes 3-43
Form number 4-60
Forms
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
FORTRAN
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
Manager
GDG see Generation
data group
Index-8
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
General Purpose Simulation System see GPSS
Generation data group 5-17
GEOMAP 7-17
Global Status Tracking
nodes 5-9
GPSS 7-24
Graphical Data Display Manager 7-21
Graphics
CalComp 7-19
CAM 7-20
CUECHART 7-15
DATA CONNECTION 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
TELLAPLAN EXPERT 7-16
Utilities 5-36
Group data sets 5-42
H
Halon system 3-8
Hardware
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
Index-9
-------
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
Graphics
Library
IMSL 7-11
In-stream procedures 4-35
defined 4-34
Information
system status 2-14
User Memos 2-14
Input only files 4-26
Input/Output devices 2-2
Inquiries
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
Reporting
System
ISPF 3-52
ISSCO-GKS 7-18
IXR see Intelligent
Transparent
Restore
Facility
Index-10
-------
NCC-IBM USER'S GUIDE
IKDEX
JAN 1988
-- J --
JARS 2-24
JCL
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
JES2
control statements 4-3, C-1
AFTER C-1
BEFORE C-2
CNTL C-2
EXCLUSIVE 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
Job
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
Index-11
-------
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
Keypunch
costs 2-23, 2-28
Keypunch Services 2-23
KWIC/KWOC 5-34
L
LABEL
on DD statement 4-24
Labels
self-sticking 4-61
Laser printer see IBM 3800
Datagraphix 9835 2-8
WIC 4-49
Librarian
Disk 2-21
Tape 2-21
Libraries
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
LRECL
on DD statement 4-13, 4-24
Index-12
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
M
Mailing
output M-54
Mailing address
National Computer Center 2-2
Manuals
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
Messages
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
Modeling
systems 7-24
MSGCLASS
on JOB statement 4-4
MSGLEVEL
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
NATURAL 7-2
NCC-IBM Users Guide
maintenance 1-2
obtaining 1-2
updates 1-2
New
accounts' 3-1
News Alerts 2-14
NOTIFY
on JOB statement 4-5
NTIS see National
Technical
Information
Service
Index-13
-------
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
OPTIMIZER III 7-23
Organization of NCC 2-1
Output
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 JCL 4-27
Output Processing 4-45
Over-the-counter service 2-2, 2-7
p
PANVAIET 7-21
Paper 4-60
Parameters
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
Password
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
Index-14
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
Physical Security 3-8
PL-MATH 7-12
PL/1
catalogued procedures 6-13
compilation 6-13
publications 6-15
PLINIT 5-36
Plotting
Anderson-Jacobson 7-21
DATA CONNECTION 7-15
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
POLYVRT 7-20
Positional parameter
on JOB statement 4-1
Power supply 2-5
Preprinted forms 4-61
Print
document quality 4-52 , 4-59
IBM 3800 4-48
PRINT control statement C-6
Print trains 4-57
Printer
costs 2-27
Printers 2-6
PRINTOFF 5-2
Priority
classes 3-14
PRTY on JOB statement 4-4
requesting 3-14
Privacy Act 3-5
Private disk packs 3-30
Problems
reporting 2-16
tape 3-49
Telecommunications 2-15, E-1
PROC see Catalogued
procedures,
see In-stream
procedures
Procedure Library-Mathematics see PL-HATH
Index-15
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
Procedure name
on EXEC statement 4-5
Procedures see Catalogued
procedures,
see In-streara
procedures,
see Private
Procedure
Libraries
Processor 2-2
PROCLIB+ 4-44
Production Control 2-19
Programming assistance 2-15
protocol converter 2-9
PRTY parameter
on JOB statement 4-4
Publications see
Documentation
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
ex
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
RECFM
on DD statement 4-12, 4-23
Refunds
for cancelled jobs 2-17
REGION parameter
on EXEC statement 4-6
on JOB statement 4-4
Index-16
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
Registration
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
Reports
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
EASYTRIEVE PLUS 7-7
IRS 7-5
Ribbon
Mylar 4-59
nylon 4-59
printing 4-59
RJE
JES2
control statements C- 1
operator commands D- 1
RJE Access 2-8, 3-55
RLSE
SPACE 4-15
Room parameter
conventions 4-55
ROUND
SPACE 4-15
Route codes 4-55
ROUTE control statement C-9
Routing
Bin ID 4-55
Plots 7-14
Remote 4-54
SYSOUT parameter 4-54
Index-17
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
S
SAS 7-9, 7-17
interactive 5-45
Schedule for NCC facilities 2-12
Scheduling
job 3-13
Scratch tapes 3-40
Scratching data set 5-16
SCRIPT 7-26
SDSF see Spool
Display
Support
Facility
Security
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
Services
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
Simulation
systems 7-24
Simulation and modeling systems
GPSS 7-24
SIMPL/1 7-25
Simulation PL/1 see SIMPL/1
Slot number 3-44
Index-18
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
Software
descriptions 2-14
summary G- 1
SORT
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
Spreadsheet
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
Storage
Central 2-2
disk 2-3
drum 2-3
tape 2-3, 3-35
Support Functions
SYNCSORT 5-33
SUPR/USER 2-16
SURFACE II 7-18
Syllabus
of courses 2-20
Symbolic parameters
coding 4-35
general 4-34
nullifying 4-36
SYNCSORT 5-33
Documentation 5-34
SYSLIB 5-30
Index-19
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
SYSLIN 5-29
SYSLMOD 5-30
SYSOUT 4-46
SYSOUT Classes 4-45
SYSOUT data sets 4-45
SYSPRINT 5-30
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 --
Tape
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
Index-20
-------
NCC-IBM USER'S GUIDE
INDEX
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
TAPESCAN 5-38
Tektronix
software 7-19
Telecommunication Service Request 2-10
Telecommunications
assistance 2-15, E-1
network 2-8
RJE Access 2-8, 3-55
Telephone Numbers A-
TELL-A-GRAF 7- 5
DATA CONNECTION 7- 5
GEOMAP 7-7
TELLAPLAN EXPERT 7- 6
Terminals
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
Management
System
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
ISO
Accessing 3-51
global status tracking 5-9
HELP 5-1
performance goals 3-50
2ED 5-1
2UEUE command 5-4
Index-21
-------
NCC-IBH USER'S GUIDE INDEX JAN 1988
Utilities 5-2, 5-4, 5-9
vs. batch 5-44
TSO/E
defined 3-49
TYMNET 2-9, 3-51, E-1
TYPRUN
on JOB statement 4-4
U --
U.S. Department of Commerce 3-2
UID 3-5
Uninterruptible Power Supply see UPS
UNIT
on DD statement 4-11, 4-21
Update
of NCC-IBM Users Guide 1-2
UPS 2-5
Usage aids 2-14
User
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
User-ID
obtaining 3-1
on JOB statement 4-2
Users Guide
updates 1-2
Utilities
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
PRINTOFF 5-2
BUEUE 5-4
SDSF 5-9
TAPESCAN 5-38
TSO 5-1, 5-2, 5-4,
5-9
Index-22
-------
NCC-IBM USER'S GUIDE INDEX JAN 1988
Utilization Identifier 3-5
V --
Virtual Storage Access Method see VSAM
VOL=SER
on DD statement 4-11
Volume-serial
of tapes 3-40
VSAM 5-32
VSAMAID/XP 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
Index-23
-------
TABLE OF CONTENTS
1.0 INTRODUCTION 1-1
1.1 PURPOSE AND SCOPE 1-1
1.2 USING THE MANUAL 1-1
1.2.1 Numbering Systems 1-1
1.2.2 Jobstreams 1-2
1.2.3 Submitting Comments 1-2
1.3 MAINTENANCE AND DISTRIBUTION PROCEDURES 1-2
2.0 GENERAL FACILITIES AND SERVICES 2-1
2.1 ORGANIZATION AND MANAGEMENT, 2-1
2.2 PHYSICAL FACILITIES 2-1
2.2.1 NCC Hardware 2-2
2.2.1.1 Processor 2-2
2.2.1.2 Central Storage 2-2
2.2.1.3 Input/Output 2-2
2.2.1.M Tape Accelerator 2-3
2.2.1.5 Controllers 2-3
2.2.1.6 Disk, Drum, and Tape Devices 2-3
2.2.1.7 Plotting and Graphic Devices 2-5
2.2.1.8 Auxiliary Devices 2-5
2.2.1.9 Computer Print and Card Read/Punch System 2-6
2.2.2 WIC Hardware 2-7
2.2.2.1 Remote Job Control Configuration 2-7
2.2.2.2 Plotting and Graphic Devices 2-8
2.2.2.3 Datagraphix 9835 Laser Printer 2-8
2.3 TELECOMMUNICATIONS NETWORK 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 SERVICES TO USERS 2-12
2.4.1 Business Hours 2-
2.4.2 User Information Services 2-
2.4.2.1 Availability of Manuals 2-
2.4.2.2 User Memos 2-
2.4.2.3 System Status Information 2-
2.4.2.4 Signon Announcements 2-
2.4.2.5 On-Line Usage Aids 2-14
2.4.3 Consultation and Assistance 2-15
2.4.3.1 Telecommunications Assistance 2-15
2.4.3.2 User Support Services 2-16
2.4.3.3 Production Control Services 2-19
2.4.3.4 Education Services 2-20
2.4.4 Data Processing Support Services (DPSS) . . . 2-21
2.4.4.1 Tape Library 2-21
2.4.4.2 Disk Library 2-21
2.4.4.3 I/O Control 2-21
2.4.4.4 Micrographics Services 2-22
2.4.4.5 Keypunch Services 2-23
2.4.4.6 Courier Service 2-23
2.4.5 Cost of Services 2-24
2.4.5.1 NCC Chargeback Algorithm 2-24
2.4.5.2 Charges for Computer-Related Services . 2-27
3.0 MANAGEMENT OF NCC-IBM SYSTEM RESOURCES 3-1
3.1 REGISTRATION PROCEDURES 3-1
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
3.1.3.1 Invalid User-ID 3-3
3.1.3.2 Invalid Account 3-4
3.1.3.3 Deleted User-ID or Deleted Account .... 3-4
3.1.3 ADP Utilization Identifier 3-5
3.2 SYSTEM SECURITY 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
IV
-------
3.2.4 Physical Security 3-8
3.2.5 Data Security 3-8
3.2.6 Disaster Recovery Plans 3-8
3.3 RESOURCE MANAGEMENT 3-9
3.3.1 Data Set Naming Conventions 3-9
3.3.1.1 Operating System (OS) Standards 3-9
3.3.1.2 NCC-IBM Conventions for User Data Sets . 3-10
3.3.1.3 NCC-IBM Conventions for Naming VSAM Data 3-10
Sets 3-10
3.3.1.4 NCC-IBM Conventions for System Data Sets 3-11
3.3.1.5 System Catalog 3-12
3.3.2 Job Scheduling 3-13
3.3.2.1 JES2 3-13
3.3.2.2 User Production 3-17
3.3.3 Direct Access Storage Device (DASD) Management 3-20
3.3.3.1 Archives 3-21
3.3.3.2 Intelligent Transparent Restore Facility 3-22
3.3.3.3 Reload From Archives 3-23
3.3.3.4 Incremental Backups 3-24
3.3.3.5 fiRELOAD's 3-26
3.3.3.6 Full Volume Saves 3-27
3.3.3.7 Scratch 3-27
3.3.3.8 Compress 3-28
3.3.3.9 Space Release 3-28
3.3.3.10 Acquiring Disk Space 3-28
3.3.3.11 Temporary On-Line Data Set 3-30
3.3.3.12 Private Disk Packs 3-30
3.3.3.13 Efficient Use of DASD at the NCC .... 3-32
3.3.4 Tape Management 3-35
3.3.4.1 Tape Data Storage 3-35
3.3.4.2 Tape Management System 3-39
3.3.4.3 Foreign Tapes 3-43
3.3.4.4 Magnetic Tape Maintenance and Disposition 3-46
3.4 TSO AND CICS ACCESS 3-49
3.4.1 TSO/E 3-49
3.4.1.1 Logon Procedures 3-51
3.4.1.2 ISPF 3-52
3.4.1.3 **LOGON 3-52
-------
3.5
3. 6
4. 0
4. 1
4. 2
U. 3
3.4.2 CICS
RJE ACCESS TO NCC-IBM
JOB ENTRY SUBSYSTEM (JES2)
PROCESSING JOBS THROUGH THE NCC-IBM SYSTEM
JOB CONTROL LANGUAGE (JCL)
4.1.1 JOB Statement
4.1.2 Execute (EXEC) Statement
4.1.2.1 Parameters
4.1.2.2 JOBLIB and STEPLIB Considerations . . .
4.1.3 Data Definition (DD) Statement
4.1.3.1 Disk Data Sets
4.1.3.2 Allocating Direct Access Storage ....
Device (DASD) Space
4.1.3.3 Tape Data Sets
4.1.3.4 Read or Input Only Files
4.1.4 OUTPUT JCL Statement
4.1.4.1 Parameters
4.1.4.2 Implicit Mode
4.1.4.3 Explicit Mode
PROCEDURES
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
4.2.5.1 Overriding EXEC Statement Parameters . .
4.2.5.2 Overriding DD Statement Parameters . . .
4.2.5.3 Adding DD Statements to a Procedure . .
4.2.6 Sample Procedure
4.2.7 User Procedures
4.2.8 Private Procedure Libraries
OUTPUT PROCESSING
4.3.1 SYSOUT Classes
4.3.2 Use of the 3800 Laser Printer
3-55
3-56
3-56
4-1
4-1
. 4-1
. 4-5
. 4-6
. 4-7
. 4-8
. 4-9
4-13
4-13
4-19
4-26
4-27
4-28
4-32
4-33
4-33
4-35
4-35
4-35
4-36
4-37
4-37
4-38
4-39
4-39
4-42
4-44
4-45
4-45
4-46
VI
-------
4.3.2.1 Forms for the 3800 4-47
4.3. 2. 2 Printing on the 3800 Printer 4-48
4.3.2.3 Job Control Language 4-49
4.3.3 Use of the Laser Printers at WIC 4-49
4.3.3.1 Job Control Language 4-51
4.3.3.2 Forms for the WIC Las__ Printers .... 4-51
4.3.3.3 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
4.3.6.1 Obtaining a Bin ID 4-55
4.3.6.2 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
5.0 SUPPORTED UTILITY FUNCTIONS 5-1
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-
5.1.7.1 Interactive Execution 5-
5.1.7.2 Batch Execution 5-
5.1.7.3 BDT Features 5-
5.1.7.4 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
VII
-------
5.2.9.1 SYSLIN 5-29
5.2.9.2 SYSLIB 5-30
5.2.9.3 SYSUT1 5-30
5.2.9.4 SYSPRINT 5-30
5.2.9.5 SYSLMOD 5-30
5.2.9.6 INCLUDE Control Statement 5-31
5.2.10 ICF Catalogs 5-31
5.2.11 VSAM 5-32
5.3 OTHER SUPPORT FUNCTIONS 5-33
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
5.3.6.1 Stacking Commands 5-39
5.3.6.2 Using MYLISTC 5-39
5.3.6.3 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
5.3.10.1 Advantages of Interactive Processing . . 5-44
5.3.10.2 Example of a CLIST 5-45
6.0 PROGRAMMING LANGUAGES 6-1
6.1 ASSEMBLER LANGUAGE 6-3
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
Vlll
-------
6.4.1 Catalogued Procedures 6-13
6.4.2 Publications 6-15
6.5 BASIC 6-15
7.0 APPLICATION PACKAGES 7-1
7.1 DATA BASE SYSTEMS 7-2
7.1.1 ADABAS Data Base Management System 7-2
7.1.2 SYSTEM 2000 7-3
7.1.2.1 Publications 7-4
7.1.2.2 DBACCESS 7-4
7.1.4 Battelle's Automated Search 7-4
Information System (BASIS) 7-4
7.2 RETRIEVAL AND REPORTING SYSTEMS 7-5
7.2.1 Inquiry and Reporting System (IRS) 7-5
7.2.2 EASYTRIEVE PLUS 7-7
7.3 MATHEMATICAL AND STATISTICAL PACKAGES 7-8
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
7.3.4.1 Catalogued Procedures 7-12
7.3.4.2 Publications 7-12
7.3.5 Procedure Library-Mathematics (PL-MATH) 7-12
7.4 COMPUTER GRAPHICS SOFTWARE 7-14
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
ax
-------
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 FILE MANAGEMENT SYSTEMS 7-21
7.5.1 PANVALET Program Management 7-21
and Security System 7-21
7.5.2 Data Catalog 2 (DC2) 7-23
7.6 DEBUGGING AND OPTIMIZING SYSTEM - OPTIMIZER III . . 7-23
7.7 SIMULATION AND MODELING SYSTEMS 7-24
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 FOCUS, A FOURTH GENERATION LANGUAGE 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
7.10 MEGACALC, AN ELECTRONIC SPREADSHEET 7-32
APPENDIX A: TELEPHONE NUMBERS A-1
APPENDIX B: SAMPLE FORMS B-1
APPENDIX C= JES2 CONTROL STATEMENTS C-1
APPENDIX D: JES2 OPERATOR COMMANDS (SP2.1.5) D-1
APPENDIX E: TELECOMMUNICATIONS INFORMATION E-1
APPENDIX F: NCC-IBM MICROGRAPHICS F-1
APPENDIX G= NCC SOFTWARE SUMMARY G-1
APPENDIX H= INSTRUCTIONS FOR PRINTING SE2UENTIAL DATA SETS . H-1
OR MEMBERS OF PARTITIONED DATA SETS H-1
APPENDIX I: ROUTE CODES 1-1
x
-------
APPENDIX J: GRAPHICS EXAMPLES (OFF-LINE) J-1
APPENDIX K= EPA/NCC TRAINING COURSE SYLLABUS K-1
APPENDIX L: INFORMATIONAL DATA SETS L-1
GLOSSARY Glos-1
XI
-------
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
Kill
------- |