Mainframe Basics & MVS

Course Mechanics
• Course Type: Self Paced or Facilitated
• Course Notes:
–If training is Facilitated, Notes are located in the Notes View.
(Please see View Master/Notes Master to see the Notes View.)
–If training is Self Paced, Notes are located in Slide View.

• This course is best viewed in Slide Show Mode.

You can download this course to your desktop (if desired) for
printing and note taking.

• Length of Course: 32 hours

• Course Audience: All who are new to Mainframe Technology

11 Nov, 2003 Introduction to Mainframe 2

 Prerequisite Training

• None

11 Nov, 2003 Introduction to Mainframe 3

 Supplementary Training

• Training on JCL, VSAM, COBOL, CICS & DB2 will

help to become a developer in the mainframe
technology.

11 Nov, 2003 Introduction to Mainframe 4

 Agenda
Classification of Computers
Mainframe Overview
MVS
Characteristics Of Mainframe
MVS – Management of User Jobs
MVS Concepts
ISPF – An Introduction
IBM Utilities – An introduction
Advanced Mainframe Concepts / Products
Summary

NOTE - Click on arrow to advance directly to the associated agenda topic.

11 Nov, 2003 Introduction to Mainframe 5
 Course Objectives

• Objective A
– To give an introduction to the concepts of
mainframes and the MVS Operating System.

NOTE - Click on back arrow to return to the Agenda.

11 Nov, 2003 Introduction to Mainframe 6

 Classification of computers:

Classification of computers:
Computer systems used for business purposes are
divided into three classes.
1. Micro Computers
2. Mini Computers
3. Mainframe Computers
Although these divisions are loosely based on the
size of the computer systems, there are no hard and
fast rules for deciding exactly where one category
ends and the next begins .In other words the
categories overlap.
11 Nov, 2003 Introduction to Mainframe 7
 Classification of computers:

Several factors determine the size of computer

system.

1. Hardware configuration
2. The nature of its application
3. The complexity of software.

Basically all computers consists of two types of

components, Processor and I/O devices.

11 Nov, 2003 Introduction to Mainframe 8

 Mainframe Overview

OS family:

In the late 1960’s there were two versions of OS in

wide spread use. They are :
1.OS/MFT
2.OS/MVT

Actually they differed in the way they handled multi

programming.

11 Nov, 2003 Introduction to Mainframe 9

 Mainframe Overview

MFT -
Stands for multi programming with fixed number of
tasks. It preallocated a fixed number of partitions
where user jobs could execute. So under MFT, the
number of jobs that could be multiprogrammed is
equal to number of preallocated partitions.

MVT – stands Multiprogramming with variable

number of tasks, the logic is just opposite to MFT.
The number of jobs that could be multiprogrammed
depends on storage available and size of partition
is decided during run time and it is variable.
11 Nov, 2003 Introduction to Mainframe 10
 MVS

MVS: (Operating System)

MVS offers multiple virtual storage that means each
multiprogrammed job is given its own virtual storage address
space which can be up to 16MB or 2GB based on our MVS
version.
MVS Versions:
 MVS/370 –Older version of MVS that limits a user’s address
space to 16MB.
 MVS/XA – New version that allows up to 2GB address space
for each user.
 MVS-ESA –latest version of MVS which let each job access
more than one 2GB address space.

11 Nov, 2003 Introduction to Mainframe 11

 MVS

O/S 390 :
IBM’s OS/390 was the next generation of
Operating Systems.
Z / OS :
Currently , this is the latest operating System by IBM.
When newer versions are released, the mainframe
operations group plans for the upgrade and
converts all the current products to the upgraded
version.

11 Nov, 2003 Introduction to Mainframe 12

 Characteristics of Mainframe

Virtual Storage:
 Virtual Storage ,which is simulated to exist by
MVS.
 Storage is not physically exist in main memory.Its
actually expansion of Real storage which is in disk
only.But always active pages will be in Main
storage and so the processing will be faster.
 Because of virtual storage ,the number of
programs that can be multiprogrammed increased.
So the productivity goes up overall though the
operating system needs to perform additional
virtual storage control functions.

11 Nov, 2003 Introduction to Mainframe 13

 Characteristics of Mainframe

Multi-Programming:
 More than one program run concurrently in the
system.(Note: anyway only one program will run at
a time)
 Improves overall productivity of system installed.
 The central component of operating system,
supervisor program which gives the processor
control to one job. When the job goes for some i/o
operation it gives the control back to supervisor.
Supervisor decides based on priorities assigned to
the jobs in queue and gives the control to highest
priority.

11 Nov, 2003 Introduction to Mainframe 14

 Characteristics of Mainframe

Spooling:
Outputs are written to disk devices and from there
it will be written into Printer when it is available.
(otherwise due to Multiprogramming concept of
MVS, various programs reports will be mixed up)
Batch Processing:
JES2/3 will take care of Priority.
Time-Sharing (Foreground Processing):
TSO- Here, user directly interacts with system.
examples - all the ISPF operations.

11 Nov, 2003 Introduction to Mainframe 15

 Configuration of Mainframe

HARDWARE CONFIGURATION

Mainframe computer system is a large collection of

computer hardware devices. Consists Mainly of
two categories:
Processors
I/O devices

11 Nov, 2003 Introduction to Mainframe 16

 Processors

PROCESSORS:
The Central components of the Mainframe Computer
Systems are the Processors. MVS runs on processors that
have evolved in the past years to enterprise systems.
The MVS runs on processor system/370, which was
developed 30 yrs back but IBM still uses them because they
have enhanced the OS capabilities significantly and in the
future also they will make it compatible to move on with the
advancements.
Basic Components are
 CPU
 MAIN STORAGE
 CHANNELS.

11 Nov, 2003 Introduction to Mainframe 17

 Processors

MULTIPROCESSING:
More than one processor was introduced to increase the
processing rate and the system availability also increases.
Here processors share access to main memory and the OS
determines how each processor is utilized. Situations where
four processors are used, operates in two modes:
As a single four processor & As two independent 2-CPU
processors.
In the latter case the main storage and the channels are
split between the two processors. Depending on the
operational needs the installation may vary,
Initially the multiprocessor installation reported to VM, an
operating system designed to emulate the multi computer
systems and configured to take advantage of the multi
computer systems.

11 Nov, 2003 Introduction to Mainframe 18

 Processors

PR/SM:
To address this problem PROCESSOR
RESOURCE/SYSTEM MANAGER was introduced.
Here the multi CPU processor was split into
several partitions. Each of this partition can operate
as an independent system. The advantage of this
system was one of these partition could serve as the
back up of the PRIMARY PARTITION. If the primary
partition fails at any point of time the back up starts
to operate from where the primary stopped. During
these situations of failures the PR/SM also
reconfigures the I/O channels assigned to the
various partitions without disrupting the processing.

11 Nov, 2003 Introduction to Mainframe 19

 IO Devices

CHANNELS:
This is the path between the processor and the I/O
devices. There are 8 channels numbered 0 thru 7, which can
communicate and pass data between the processor and the
devices.
Each channel can connect up to eight devices called as
CONTROL UNITS. Depending upon the Processor and
device, a control unit can be fixed within the processor’s
cabinet or device cabinet or in it’s own cabinet.
A channel itself is a small computer in the sense that it
executes I/O instructions called Channel Commands that
operate the I/O devices that are attached to it. As a result the
channel frees the processor to execute other instructions.
Since channel processing overlaps CPU processing, overall
system performance improves.

11 Nov, 2003 Introduction to Mainframe 20

 IO Devices

The main purpose of these devices are to provide

with input, receive output and provide secondary
storage. The common types of I/O devices are

Unit record devices

Magnetic tape devices
Direct access devices &
Telecommunication devices

11 Nov, 2003 Introduction to Mainframe 21

 IO Devices

Unit record devices:

These are two types of Unit record devices. They are card
devices and printers. Each record processed by the device is
a physical unit hence the name “Unit record devices”.
Incase of card devices, each record is a punched card.
Similarly for a printer it is a printed line.
Card devices are not commonly used anymore are of three
types. They are:
 Reader
 Puncher
 Reader/Punches.

11 Nov, 2003 Introduction to Mainframe 22

 IO Devices

Impact Printers:

Produce printed output by striking an image of

characters to be printed against ribbon, which in
turn transfers ink to the paper. The most common
type of impact printers uses a train of characters
that spins at a high speed; when the correct
character passes a print position, a hammer a
strikes the character against the ribbon to produce
the printed text. Most impact printers operate in the
range of 600 to 2000 lines per minute.

11 Nov, 2003 Introduction to Mainframe 23

 IO Devices

Non-Impact Printers:
These printers use Laser Technology to print text and
graphic images. The IBM 3800 printing Subsystem
can print at rates of up to a remarkable 20000 lines
per minute. The actual speed of this system
depends on the size of each page and the number
of lines per inch, since the mechanism involved
can transfer images to the paper an entire image at
a time.

11 Nov, 2003 Introduction to Mainframe 24

 IO Devices

For a standard size paper (11x14) and

normal print size (6 lines per inch), the 3800
can print 10200 lines per minute. At this
transfer rate the 3800 can process more
than a mile and a half of paper each hour.

11 Nov, 2003 Introduction to Mainframe 25

 IO Devices

Magnetic Tape Devices:

A tape drive reads and writes data on a magnetic tape
that’s a continuous strip of plastic coated on one side with
metal oxide. Tape drives process tapes that’s sealed within a
special type cartridge but earlier most tape drives processed
tape wrapped around an open reel much like an old
fashioned reel to reel tape recorder.
The amount of data a reel or cartridge of tape can contain
depends on the length of the tape and the density used to
record the data. Density is a measurement of how many
bytes are recorded in one inch of tape. Tape densities for
standard reel tapes are usually 1600 or 6250 bytes per inch.
Cartridge tape drives can record data using much higher
densities.

11 Nov, 2003 Introduction to Mainframe 26

 IO Devices - Magnetic Tape Devices (Cont.)

Data records are normally written to tape in-groups

called blocks, and the empty areas are called as gaps.
These separate the blocks from one and other. The
larger the block, better the usage of tapes and their
efficiency. But one drawback is that, A buffer is required
is required to store these blocks in the main storage.
The other major drawback is that the records have
to be stored sequentially. To read the 50000th record
on a tape the 49000 records have to be read to reach
that particular record. So direct access cannot be
performed here.

11 Nov, 2003 Introduction to Mainframe 27

 IO Devices - Magnetic Tape Devices (Cont.)

The tape drives are used for offline

storage of large quantities of data. To
attach a tape drive to a processor a control
unit is required. For some models the
control unit is inside one of the tape drives
and for the other models it is in a separate
cabinet.

11 Nov, 2003 Introduction to Mainframe 28

 IO Devices

Direct Access Drives:

The official IBM term for a disk drive is direct access
storage device or DASD.
These allow direct and rapid access to large amount of
data. They have become a key component of mainframe
systems. They are not only used to store user programs and
data but also to store programs and data for operating
systems functions.
Disk drives read and write data on a disk pack (also
called as volume). A disk pack is a stack of metal platters
coated with metal oxide material. Data is stored on both the
sides of the platters.
Earlier removable disk pack’s were used but now
permanent disk pack’s are used. The advantages of fixed
disk packs are that they are more reliable and faster.

11 Nov, 2003 Introduction to Mainframe 29

 IO Devices

Tracks & Cylinders:

Data is recorded on the usable surfaces of a disk
pack in concentric circles called tracks.
The number of tracks per surface varies with each
device type. For example a disk pack consisting of 19
usable surfaces, each with 808 tracks has a total of 15352
tracks.
Access mechanism or Actuator is the component that
reads or4 writes data on the tracks of a disk pack. The
actuator has one read/write head for each recording
surface. When the actuator moves all of it’s head move
together so they are all positioned at the same track of
each recording surface. As a result the disk drive can
access data on all those tracks without moving the
actuator.

11 Nov, 2003 Introduction to Mainframe 30

 IO Devices - Tracks & Cylinders (Contd.)

The tracks that are positioned under the

heads of the actuator at the one time make
up a cylinder,
As a result, there are as many tracks in
a cylinder as there are usable surfaces on
the pack.

11 Nov, 2003 Introduction to Mainframe 31

 IO Devices

Device capacity and data format:

There are two basic types of disk drives that store
their data in different formats. They are count key data
devices where data is stored in variable length blocks
and the other format is called as fixed block
architecture, stores data in fixed length blocks which is
not supported by MVS.
So we will deal only about CKD devices, here a
count area and a key area precede each block. The
disk revolves counter clockwise, the read/write head
encounters the count and key areas before the data
area. The count area contains the information needed
to locate and process key and data areas.

11 Nov, 2003 Introduction to Mainframe 32

 IO Devices - Device capacity and data
format (Contd.)

One of the problems with CKD devices is

that the data capacity of each track depends
on the size of the blocks used to store the data.
That’s because gaps required to separate the
count, key and data areas. Just as gaps are
required on magnetic tapes. The data
efficiency depends on how the blocks and gaps
are used. When more blocks are stored, more
gaps are also used so the total capacity
reduces.

11 Nov, 2003 Introduction to Mainframe 33

 IO Devices

Control Units:
Each type of DASD device requires two kinds of
control units to attach it to a processor channel.
The first called string controller, attaches a group of
DASD’s of the same type, the resulting group called string.
The number of devices that can be connected on one
string depends on the device type. For the3 3390 model
around 32 drives can be connected to a string.
The second type of control unit called as string
controller connects up to eight strings of DASD units to a
channel. The most common type of storage control is the
3990, which attaches two strings. If both the strings are
3990 drives up to 64 drives can be connected.

11 Nov, 2003 Introduction to Mainframe 34

 IO Devices - Control Units (Contd.)

The 3990 storage control provides high-speed cache

storage that acts as a buffer between the processor and disk
units. Special circuit keeps track of what data is accessed
most frequently and tries to keep that data in the cache
storage. When data is accessed again, it directly loads the
data from cache than accessing the DASD. Depending on
the model the size of the cache can range from 32mb to
1024mb & improves the system’s overall performance.
In addition the 3990 storage control also provides the
facility of connecting to more than one channel to the
processor. This enables several simultaneous disk
operations to be processed at one instance. The smallest
3990 models support up to four standard channel
connections and the largest can support up to 16 standard
connections or 128 ESCON channel connections.

11 Nov, 2003 Introduction to Mainframe 35

 IO Devices

Data communications equipment:

Lets an installation create a data communications
network that lets the users at local terminal and remote
terminal to access the system.
Elements of a data communications network are:
 The Host
 Communications Controller
 Modems
 Telecommunications Lines &
 Terminal Systems

11 Nov, 2003 Introduction to Mainframe 36

 IO Devices

The center of the network is the host system

and the processor. The control unit that attaches
to the host system’s channels is called the
communication controller. This manages the
communication functions necessary to connect
remote terminal systems via Modems and
Telecommunication Lines.
If the terminal system is located close
enough to the host system, the Modems and
Telecommunication Lines can be eliminated.
Then, the terminal can be connected directly to
the communications controller or one of the host
processor’s channel.

11 Nov, 2003 Introduction to Mainframe 37

 MVS Management of user jobs:
JOB:
Job is an execution of one or more related programs or
procedures in sequence. Each program or procedure is called a
job step. E.g. To get a report of an organization’s employees in
alphabetic order, first program should sort the file by name and
the second program prints the report. Here, one job has 2 steps.
JOB Life cycle:
1.The job is submitted
2.The job is selected for execution.
3.The job is executed
4.The job’s output is processed.
5.The job is purged.

11 Nov, 2003 Introduction to Mainframe 38

 MVS Management of user jobs:

Job Entry Subsystem:

This MVS component keeps track of
jobs that enter the subsystem, presents
them to MVS for processing and sends their
spooled output to correct
destination.(printer)
Original version of OS did not provide
this facility. It has job scheduler ,a relatively
crude form of present JES was totally
inadequate most installation’s need.

11 Nov, 2003 Introduction to Mainframe 39

 MVS Management of user jobs:

Houston Automatic spooling program was used

which ran as a job in scheduler with most of the
functions of job scheduler and provided most efficient
operation and comprehensive control than the job
scheduler alone.(HASP)
Asymmetric multiprocessing system (ASP) was
used in shops which had more than 1 processor. It
gave good control of job processing in multiprocessing
environment.
When MVS was announced, IBM integrated the
functions provided by HASP and ASP into operating
system by providing the JES. But since both were
incompatible with each other, two subsystems were
created .JES2 for HASP and JES3 for ASP.

11 Nov, 2003 Introduction to Mainframe 40

 MVS Management of user jobs:

How JES2/JES3 process jobs:

Job entering into subsystem: By SUBMIT

command in the jcl .JES2 (internal reader) read the job
stream from the DASD file and copy it to job queue,
which is a part of special DASD file called JES spool.
Job selected for execution :
Based on class and priority jobs are selected for
execution. Job classes are assigned based on the
processing characteristics of the job. job will execute
within 15 or 30 minutes etc. of submission. These
definitions will be done at installation time.

11 Nov, 2003 Introduction to Mainframe 41

 MVS Management of user jobs:

Initiator is program that runs in the system region of

address space that’s eligible for batch job processing.
Each initiator can handle one job at a time.
1– A
2– B,C
3– C,D,E
The number of active initiators on a system and as
a result number of address spaces eligible for batch job
processing determines the number of batch jobs that
can be multi programmed at once. Initiators can be
started when MVS is activated and that can be stopped
or started by an operator while MVS is running.

11 Nov, 2003 Introduction to Mainframe 42

 MVS Management of user jobs:
Job Execution:
The first thing an initiator does after selecting a job for
execution is to invoke a program called interpreter. The
interpreter’s job is to examine the job information
passed to it by JES and create a series of control
blocks in the scheduler work area(SWA), a part of
address space ‘s private area. These control blocks
describe all of the datasets the job needs.
After the job creates the SWA blocks the initiator
goes through three phases for each step in the job.

Allocation routines:
Allocate all the required datasets needed.
11 Nov, 2003 Introduction to Mainframe 43
 MVS Management of user jobs:

User region:
The initiator builds a user region where user
programs can execute, loads the program into that
region and transfers control to it. As the program
executes it uses the control blocks for the resources
allocated to it.
Un allocation routine:
When program is completed, the initiator invoke
this which release any resource used by job step.
Based on output class, the processed outputs are
sent and the job is purged from the system. Simply put,
that means JES spool space used by the job freed so it
can be used by other jobs. Any JES control blocks
associated with that job are deleted.
11 Nov, 2003 Introduction to Mainframe 44
 Data Management By MVS:

Dataset:
Collection of related data that is managed
as MVS as a unit. Within the dataset, data is
organized into smaller units called records,
which can be processed individually by
application programs.
Label Processing:
When dataset is normally stored on disk or
tape MVS normally identifies it with special
records called labels. To access a dataset
correctly, the label information passed in the
JCL and what is there in MVS should match.
11 Nov, 2003 Introduction to Mainframe 45
 Data Management By MVS:

DASD labels:
Volume label : Unique six char volume number –
cylinder 0 track 0 record 3 contains volume number and
pointer to VTOC.
VTOC : contains five format of records.
FORMAT-4 information abt VTOC itself.
FORMAT-1 pointers to dataset which has maximum of
3 extents(1 primary,2 extents) information.
FORMAT-3 pointers to dataset which has the remaining
13 extents
FORMAT-2 ISAM files
FORMAT-5 free extent pointers(26 extent-1)
11 Nov, 2003 Introduction to Mainframe 46
 Data Management By MVS:

Tape Label:
Unlike DASD, for tape labels are optional. Here
labels are stored along with files rather than
separately. tape file started with HDR1 which is
followed by files data and at the end EOF1 end-of-
file label exist.
Catalogs:
Many MVS installations have hundreds of
DASD volumes and thousands of datasets. So it is
impossible to remember the DASD for accessing
dataset and we cannot give Volser for each dataset
we want. So cataloging is necessary.
11 Nov, 2003 Introduction to Mainframe 47
 Data Management By MVS:

There are two types of catalogs:

One master catalog and unlimited number of user catalogs.
Master catalog:
1.System dataset entries
2.User catalog entries
Importance of high level qualifier:
The high-level qualifier and the user catalog name are the
same. More often the high level qualifier will be alias of the
actual name.(alias is simply alternate name for something, in
this case user catalog)
Fst-qualifiercataloguser catalogdasdvtoc
dataset

11 Nov, 2003 Introduction to Mainframe 48

 Data Management By MVS:

Dataset organization:

Within a dataset, data can be organized in one of

several ways depending on how the data will ultimately
be processed.

NON-VSAM:
1. Sequential
2. Indexed sequential
3. Direct
4. Partitioned

11 Nov, 2003 Introduction to Mainframe 49

 Data Management By MVS:

Depends on dataset purpose, file organization is

selected.
For example, in every day transaction only very few
records will be updated and they are random means
then Indexed sequential or Direct is the best. For payroll
file Sequential is well enough.
Dataset processing:
Allocation at 3 levels:
1.Unit is selected and allocated (SYSDA,TAPE)
2.Volume is allocated (VOL=SER= )
3.Dataset on that volume allocated.
(VTOC updated in case of disk)
11 Nov, 2003 Introduction to Mainframe 50
 Data Management By MVS:

There are 3 access methods available:

 Basic ( BSAM,BISAM,BDAM)
 Queued (QSAM,QISAM)
 VSAM (KSDS,ESDS,RRDS)

When the file is opened in program:

The relation between program, dataset and access
method has to be established. This connection is
established through a special control block called
DCB. Similarly for VSAM it is ACB.

11 Nov, 2003 Introduction to Mainframe 51

 Data Management By MVS:

DCB,ACB are basically tables where the vital information

about status of dataset is stored as it is processed.
OPEN : Initialize this control block and information about
dataset comes from JCL/Program/catalog/label.

I/O requests: Access methods used for accessing required

records.
CLOSE : File is disconnected from the program.
DEALLOC : Files are automatically deallocated when the
job is finished with it.

During deallocation disposition is the key thing which

decide what to do with that file.

11 Nov, 2003 Introduction to Mainframe 52

 Data Management By MVS:

Storage Management Subsystem:

 New Feature of MVS
 Removed many of the manual procedures that are
associated with managing datasets, such as which
volume the dataset should be stored on, amount of
space to allocate the dataset, when the dataset no
longer need and can therefore be deleted or moved to
offline storage.
 Based on job name ACS routine assign the
corresponding data class, storage class, management
class for the dataset.

11 Nov, 2003 Introduction to Mainframe 53

 MVS Concepts
VIRTUAL STORAGE:
This is a facility that simulates a large amount of main storage
by treating DASD storage as an extension of real storage . In other
words when virtual storage is used, the processor appears to have
more storage than it actually does.
ADDRESS SPACE:
To refer to a particular location, we need to use an address
that indicates the storage location’s offset from the beginning of
memory. The first byte of storage is at address 0, the second byte at
2. etc .
Address space is simply the complete range of addresses and
as a result the number of address locations that can be accessed
by the computer.
The maximum size of address space is limited by the number
of digits that can be used to represent address.
11 Nov, 2003 Introduction to Mainframe 54
 MVS Concepts

MULTIPLE VIRTUAL STORAGE:

 MVS not only simulates more storage but also uses
real storage to simulate several address spaces,
each of which is independent of the others . In fact
MVS derives its name from this technique (Multiple
virtual storage)
PAGING:
 We know real storage is very low and that raised the
need for virtual storage. Implementation of virtual
storage is possible through Paging concept.
Lets see how it works.

11 Nov, 2003 Introduction to Mainframe 55

 MVS Concepts

 Virtual storage divided into 4K pages. Real storage divided

into 4K page frames. DASD area used for virtual storage
called page dataset is divided into 4K page slots each of
which holds one page of virtual storage.

 Page fault , Page-in, Page-out:

When a program refers to a page that is not in real
storage page fault occurs. When that happens MVS
locates the page that contains the needed data on DASD
and transfers it into real storage. The least used page is
released from real storage to page dataset (page-out) and
new page put into real storage.

11 Nov, 2003 Introduction to Mainframe 56

 MVS Concepts

Usually page frames in real storage contain

pages from more than one address space. MVS keep
track of what pages are in what page frames by
maintaining tables that reflect the current status of
real storage and of each address space. As long as
the corresponding address space is active, the
corresponding tables cannot be paged out. They
should be in real storage.

11 Nov, 2003 Introduction to Mainframe 57

 MVS Concepts

EXPANDED STORAGE:
This storage acts as a large buffer between real
storage and Page dataset. During page out, the
pages are moved out to expanded storage. This
transfer occurs at CPU speed and not DASD speed
and so the operation is almost instantaneous. Pages
are written into actual page datasets when the
expanded storage becomes full.

11 Nov, 2003 Introduction to Mainframe 58

 MVS Concepts

SWAPPING:
This is same as paging but at a higher level.
Rather than move small 4K pieces of virtual storage
in and out of real storage , swapping effectively
moves entire address space in and out of virtual
storage. Because paging occurs only for address
spaces that are currently in virtual storage , paging
does not occur for address spaces that are swapped
out.

11 Nov, 2003 Introduction to Mainframe 59

 MVS Concepts

During swap out, the address spaces’

critical pages – the ones that contain the tables
that keep track of the location of each virtual
storage page for address space are written to a
special dataset called swap dataset. Later when
the system can accommodate the job again ,
the address space is swapped in so it can be
processed again.
PROGRAM MODES:
Real mode: Programs that cannot be paged or
swapped. The programs which are responsible
for virtual storage functions and tracking should
be in main storage always.
11 Nov, 2003 Introduction to Mainframe 60
 MVS Concepts

 Critical time dependant channel devices

communication programs also non –page able. Most
of the programs actually running in virtual mode only.
DATA WITHIN EACH ADDRESS SPACE :
SYSTEM AREA – NUCLEUS/ COMMON AREA:
 Operating system programs and data are in system
area and common area. This is common and shared
by all. That is system area and common are the same
for each address space.
 This controls the operation of paging and swapping of
virtual storage. System area operates in real mode.

11 Nov, 2003 Introduction to Mainframe 61

 MVS Concepts
 Three important operating system components are in
common area.
SQA: SYSTEM QUEUE AREA :
Important system tables and data that are used by
programs residing in the nucleus (Non-Pageable)
CSA: COMMON SERVICE AREA :
Similar to SQA but pageable
PLPA:
Pageable Link Pack Area: contains some operating
system programs that don’t have to be fixed

11 Nov, 2003 Introduction to Mainframe 62

 MVS Concepts

This is the portion of address space that contains data that are
unique for each address space. Within each jobs private area,
System region: Area of storage used by operating system programs
that provide services for user programs running in the private area.
At the top of private area, three local system areas that contain
information that applies only to the private area of particular address
space.
LSQA: Local System Queue Area: Tables used to control the
private area , including the tables needed to manage the private
area’s virtual storage. During swap out, this is what written into swap
dataset.

11 Nov, 2003 Introduction to Mainframe 63

 MVS Concepts

Scheduler work area:

SWA tables used to manage the execution of jobs and
programs within the private area.
Sub Pool : additional system information.
3.User Region:
This is the place where the program actually execute.
The size of user region varies depending on the amount
of storage required by the program being executed . (Max
10 MB to 12 MB) Remaining space is unallocated
storage.

11 Nov, 2003 Introduction to Mainframe 64

 MVS
MVSConcepts
Concepts

MVS/ESA DATA SPACES AND

HYPERSPACES
 Here one user can have access to more than one
address space. Depending on the requirements of
application, the additional address spaces created for
data storage can be one of the two types :data spaces
or hyperspaces
 Data spaces resides in normal storage and are
subject to paging and swapping and they are
managed by user-written programs.
 Hyperspaces are in expanded storage and managed
by MVS/ESA.
11 Nov, 2003 Introduction to Mainframe 65
 MVS
Classification
MVS Concepts
Conceptsof computers:

SYSTEM GENERATION AND INITIALIZATION:

 System generation and initialization are activities that
are required to establish a working MVS system.
 System generation is the process creating an MVS
system, and system initialization is the process of
starting a previously generated MVS system.
 Both of the above are the responsibility system
programmers staff.

11 Nov, 2003 Introduction to Mainframe 66

 MVS
Classification
MVS Concepts
Conceptsof computers:

SYSTEM GENERATION:
For the installation of the MVS system, IBM sends
the basic components that make up MVS on a series of
tapes called distribution libraries.
System generation is only a part of the overall
process of installing MVS from the distribution libraries,
selects and assembles the various components an
installation needs to create a working MVS system. To
control System generation, often called sysgen, a
systems programmer codes special macroinstructions
that specify how the MVS components from the
distribution libraries should be put together.
11 Nov, 2003 Introduction to Mainframe 67
 MVS
Classification
MVS Concepts
Conceptsof computers:

Interestingly, an installation must already have an MVS

system is required before it can generate a new one.
That’s because an existing MVS system is required to
execute the sysgen macro instruction.
Fortunately, most installations perform sysgen to upgrade
to a newer version of MVS or to make changes to their
current version. So they can use their current version of
MVS to execute the sysgen.
For new installations that don’t already have an MVS
system, the system installation process includes setting
up a small, limited function MVS system that can execute
sysgen for the complete, full function MVS system.
11 Nov, 2003 Introduction to Mainframe 68
 MVS Concepts

SYSTEM INITIALIZATION:
 Once a MVS system has been generated, it can
be used to control the operation of the computer
system.
 To begin a system initialization, the system
operator uses the system console to start an initial
program load or IPL. That causes the computer
system to clear its real storage and begin the
process of loading MVS into storage from the
system libraries and ready to process your work.

11 Nov, 2003 Introduction to Mainframe 69

 MVS Concepts

Initialization options come from one of two sources:

 The system operator or a special system library

called SYS1.PARMLIB.
 By specifying options in SYS1.PARMLIB, MVS can
initialize with little operator intervention.

11 Nov, 2003 Introduction to Mainframe 70

 MVS Concepts

Here are few examples of the system data sets:

 SYS1.NUCLEUS
 SYS1.PARMLIB
 SYS1.LINKLIB & SYS1.LPALIB
 SYS1.MACLIB
 SYS1.PROCLIB
 SYS1.CMDLIB

11 Nov, 2003 Introduction to Mainframe 71

 MVS Concepts

SUBSYSTEM AND OTHER FACILITIES:

 A complete production MVS system contains a
variety of other software products. Subsystems are
those software products, which operate in their
own address spaces under the control of MVS
system. What ever they do in that address space is
of no concern for the MVS. These subsystems
provide services that duplicate services provided
by the operating system.
 Multiprogramming, job management and spooling
are few examples of the services that are provided
by the subsystem, which are duplicated.

11 Nov, 2003 Introduction to Mainframe 72

 MVS Concepts

Some of the examples are:

 JES2/JES3
 TSO & ISPF
 TELECOMMUNICATIONS
 CICS
 IMS
 DB2

11 Nov, 2003 Introduction to Mainframe 73

 ISPF
INTRODUCTION

11 Nov, 2003 Introduction to Mainframe 74

 ISPF - Introduction

What is ISPF?
The Interactive System Productivity
Facility is a Dialog Manager that
provides tools to improve program, dialog
and development productivity and
control.

11 Nov, 2003 Introduction to Mainframe 75

 ISPF - Introduction

What the ISPF Editor does?

‘Edit’ Option (usually 2) . It is used for
create, display and change data stored in
ISPF libraries or other PDS or PS.

11 Nov, 2003 Introduction to Mainframe 76

 Line Commands

 Insert or Delete lines

 Repeat lines
 Rearrange lines or Overlay portions of lines
 Simplify text entry and formatting
 Define an input mark
 Shift data
 Include or exclude lines from the display
 Control tabs and boundaries for editing
 Convert some types of special temporary lines
to data lines

11 Nov, 2003 Introduction to Mainframe 77

 Primary Commands
 Control your editing environment.
 Find a specific line.
 Find and change a character string.
 Combine several members into one.
 Split a member into 2 or more members.
 Submit data to the job stream.
 Save the edited data or cancel without saving.
 Sort data.
 Delete lines.
 Run an edit macro.
11 Nov, 2003 Introduction to Mainframe 78
 Edit macros
• Edit Macros are ISPF Dialogs that run in
the PDF editor Environment.

• Edit Macros are nothing but the line or

primary commands or combination of
line and Primary commands.

11 Nov, 2003 Introduction to Mainframe 79

 Line Commands

 I Insert
 R Repeat
 M Move
 D Delete
 C Copy
 TE Text Edit
 TS Text Split

11 Nov, 2003 Introduction to Mainframe 80

 Line Commands

 UC - Upper Case
 LC - Lower Case
 CC - Block Copy
 DD - Block Delete
 MM - Block Move
 (( - Shift Left
 )) - Shift Right

11 Nov, 2003 Introduction to Mainframe 81

 IBM Utilities
INTRODUCTION

11 Nov, 2003 Introduction to Mainframe 82

 Introduction to IBM Utilities

 What is an utility?
 What are the types of utilities?
 What can you do with utilities?
 How to use them?
 Some examples of utility programs

11 Nov, 2003 Introduction to Mainframe 83

 ANY TYPICAL UTILITY JCL

//JOBNAME JOB (ACCOUNTING INFORMATION),

// PROGRMAMER NAME,
//STEPNAME EXEC PGM=UTILITY-NAME,
// PARM=PARM-VALUE
//PRINTNAME DD SYSOUT=PRINT-DEVICE-CLASS,
//INPUT-FILE DD INPUT-FILE-FEATURES
//OUTPUT-FILE DD OUTPUT-FILE-FEATURES
//WORKFILE DD WORK-FILE-FEATURES
//INPUT-DATA DD *
***********
/*
11 Nov, 2003 Introduction to Mainframe 84
 IEBCOPY

 How to use them?

 Copy a PDS
 Compress a PDS
 Merge PDSs
 Selecting or excluding specified
members during copying or merging

11 Nov, 2003 Introduction to Mainframe 85

 A SAMPLE IEBCOPY
//MYCOPY01 JOB (J007,,,),'IEBCOPY’,CLASS=U,MSGCLASS=Y,
// TYPRUN=SCAN,NOTIFY=&SYSUID
//COPY01 EXEC PGM=IEBCOPY
//SYSPRINT DD SYSOUT=*
//SYSUT1 DD DSN=INPUT.PDS.TOBE.COPIED,DISP=SHR
//SYOUT2 DD DSN=OUTPUT.PDS.WHICH.ISTOBE.CREATED,
// DISP=(NEW,CATLG,DELETE),UNIT=3390,VOL=NKP900,
//SYSIN DD *
COPY INDD=SYSUT1,OUTDD=SYSUT2
SELECT MEMBER=(MEM1,MEM4,MEM7)
/*
//

11 Nov, 2003 Introduction to Mainframe 86

 IEFBR14
It’s a two line program
 It simply clears register15, then passing a return code 0,
and then branches to the address in the register 14, which
returns control to the System
 Why is it a do-nothing program?
 What you can do with IEFBR14?
- Scratch a Dataset
- Catalogue and Uncatalog a dataset
- Careate and Delete datasets
- As a testing utility

11 Nov, 2003 Introduction to Mainframe 87

 IEBCOMPR

 Compare sequential datasets

 Compare partitioned datasets

11 Nov, 2003 Introduction to Mainframe 88

 IEHLIST

 List the contents of partitioned

data set directory
 List entries in a VTOC
 List entries in a system catalog

11 Nov, 2003 Introduction to Mainframe 89

 IEHPROGM

 To modify system control data in VTOC,

data-set label, and the system catalog
 Scratch/Rename a member of a PDS
 Catalog or uncatalog a dataset
 Build an index for GDG

11 Nov, 2003 Introduction to Mainframe 90

 IEBGENER

 Copy a sequential file

 Reorganize records in a sequential file
 Change the record or block length of a
sequential file
 Print the contents of a sequential file
 Creating a library member

11 Nov, 2003 Introduction to Mainframe 91

 Advanced Mainframe Concepts
/ Products

INTRODUCTION

11 Nov, 2003 Introduction to Mainframe 92

 Advanced Mainframe Concepts / Products

Datasets
JCL
JES2 / JES3
SDSF
CA-7
CA-VIEW
IBM Fault Analyzer
Change Management
ENDEVOR
11 Nov, 2003 Introduction to Mainframe 93
 DATASET

A data set is a named collection of data that

contains individual data units organized (formatted)
in a specific, IBM-prescribed way and accessed by
a specific access method that is based on the data
set organization.
Types of data set organization:
 Sequential
 Relative sequential
 Indexed sequential and
 Partitioned

11 Nov, 2003 Introduction to Mainframe 94

 Types of Datasets

Physical Sequential
A Physical Sequential file, DSORG=PS, is a
simple file with records stored in the order that they
are written. PS files are typically used for text and
logs. Large PS files which are only ever required by
one task at a time, are very suitable for tape. It is
possible to improve the performance of PS files by
striping them over several volumes, but this is only
useful if they are used by concurrent tasks

11 Nov, 2003 Introduction to Mainframe 95

 Types of Datasets (Contd.)
Partitioned
The minimum size of data allocation under OS/390 is 1
track, or 56,664 bytes. This means that a Physical Sequential
file which contained three 80 byte records will contain 240
bytes, but occupy 56,664. A Partitioned Dataset (PDS) solves
this problem by combining a lot of small files into one large
container. The individual files are stored as members within
the PDS. Each member must have a unique 1-8 character
name.
Issues in Partitioned datasets:
There are a number of issues with PDS files. When you
delete a member, the space it occupied is not reused. This
means that over time, the PDS will grow, and eventually fill
up. The easiest way to fix this, is to list the file on ISPF option
3.4, then enter a 'Z' against it. This will compress the file, and
all the deleted space will available for use.
11 Nov, 2003 Introduction to Mainframe 96
 Types of Datasets (Contd.)

VSAM Structures:
There are 4 types of VSAM Dataset.
 Key Sequence Data set
 Entry Sequence Data Set
 Relative Record Data Set
 Linear Data Set.
Virtual Sequential Access Method is discussed
in the VSAM Section.

11 Nov, 2003 Introduction to Mainframe 97

 GDG (Generation Data Group )

Generation Data Group.

Collection of (z/OS non-VSAM) datasets with
the same logical name (GDG Base Entry).
The individual datasets are uniquely identified
by the generation number which is stored as part of the
dataset name.
The datasets can be referenced either by the
explicit generation number or the relative generation
number. GDGs are useful where datasets are cycled –
standard JCL can be used without having to change the
dataset names for each run.

11 Nov, 2003 Introduction to Mainframe 98

 JCL (Job Control Language)

JCL:
Job Control Language, or JCL, is a
set of control statements that provide the
specifications necessary to process a job,
and the series of JCL statements that make
up a job is typically called a job stream.

11 Nov, 2003 Introduction to Mainframe 99

 JES2 & JES3

A job is the execution of one or more related programs

in sequence. Each individual program executed in a job is
called a job step.
To process a job, a Job Entry Subsystem, or JES, is
used. It keeps track of jobs that enter the system, presents
them to OS/390 for processing, and sends their spooled
output to the correct destination.
There are two versions of JES available under OS/390:
JES2 and JES3. Since JES2 and JES3 are incompatible
with one another, your shop should have only one or the
other installed.

11 Nov, 2003 Introduction to Mainframe 100

 JES2 & JES3

How JES2 and JES3 process jobs :

Job is Job is Job is Job’s Job is

submitted selected executed Output is purged
for processed
execution

How a job is entered into the system

Jobs are created by entering JCL commands into a display
terminal. Then, the jobs are stored in files on DASD.
To enter, or submit, the job into the system, the terminal
user issues a SUBMIT, or SUB, command. That causes JES2 or
JES3 to read the job stream from the DASD file and copy it to
the job queue on the JES spool.

11 Nov, 2003 Introduction to Mainframe 101

 SDSF (System Display and Search Facility)

System Display and Search Facility:

On-line tool for programmers and operators
monitoring jobs awaiting execution in the JES2 input spool
queues and, most commonly, viewing the printed output of
batch jobs in the Held output spool queues, to save
printing it on paper.
Most users can only view their own jobs,
but systems programmers and operators are normally
allowed to view everything, making SDSF especially useful
for monitoring jobs currently executing, or the rest of the
output of that seemingly endless print job. Runs in ISPF or
directly in TSO without ISPF. An optional, separately
priced feature of z/OS.

11 Nov, 2003 Introduction to Mainframe 102

 CA - 7

What is CA-7?
CA-7 is a Production control system. It’s an online, real-
time, interactive system which automatically controls, schedules
jobs based on the Date and time, job dependencies and
available resources.
Functions of CA-7
 Provides Online Scheduling facility
 Selects job for execution
 Submits jobs
 Tracking jobs
 Analyzes job execution results

11 Nov, 2003 Introduction to Mainframe 103

 CA - View

To view the output of the Job in online. Online

viewing has been implemented to significantly
reduce the amount of microfiche produced by
Information Services. Instead of producing
microfiche, archival copies of production reports will
be redirected to an online viewing tool called CA-
View.

11 Nov, 2003 Introduction to Mainframe 104

 CA – View (Contd.)

Online viewing represents a significant improvement

over traditional microfiche for the following reasons:
 reports are available immediately; no waiting for fiche
delivery
 online viewing has full search capabilities
 reports are more secure; only authorized users are
permitted to view reports
 automatic records retention; CA-View will
automatically delete unnecessary reports after the
amount of time you specify

11 Nov, 2003 Introduction to Mainframe 105

 IBM - FAULT ANALYSER

 IBM Fault Analyzer for z/OS and OS/390 assists the

developer in quickly analyzing and fixing application and
system failures.
 Helps determine the cause of failure while further providing
assistance in how to resolve the problem.
 Provides support for CICS® system abends, as well as
MQSeries® application abends.
 Supports different modes of operation:
– Real-time analysis at time of abend
– Batch re-analysis of faults in history file
– Interactive re-analysis of faults in history file under ISPF

11 Nov, 2003 Introduction to Mainframe 106

 IBM - FAULT ANALYSER (Contd.)

 Facilitates expansion of messages and abend codes

applicable to the analysis result.
 Lets an application add its own application specific messages
and abend codes to supplement those supplied by IBM.
 Provides management of application abends; the fault history
file helps track and manage application abends, fault reports,
and associated dumps through an interactive display.
 Eliminates need to recompile programs or change JCL to
invoke IBM Fault Analyzer for z/OS and OS/390®.
 Applies the collective knowledge of many experts to each
abend.
 Offers interactive re-analysis to provide programmers with the
ability to increase the level of detail the analysis engine can
provide after the abend has occurred.

11 Nov, 2003 Introduction to Mainframe 107

 CHANGE MANAGEMENT

In the mainframe world, there are many people in a

given project group. To ensure that a module /
program / JCL member etc. is not changed
inadvertently , we have change management
teams and products in place.
These personnel, with the help of the Change
management products, manage various versions of
a module and account for its proper movement
through various stages like test, QA, production
etc.
11 Nov, 2003 Introduction to Mainframe 108
 ENDEVOR

What is Endevor?
ENDEVOR is a Change
Management software product
that controls, automates, and
monitors the entire application
development life cycle.
ENDEVOR allows you to
automate and control the
movement of source code from
development through to
production (see diagram).

11 Nov, 2003 Introduction to Mainframe 109

 ENDEVOR (Contd.)

ENDEVOR gives you the ability to:

 Automatically compare and track your changes against the
PRODUCTION version, creating an on-line change history.
This speeds up the debugging process and enables you to
see exactly WHO changed it, WHAT was changed, WHEN it
was changed, WHERE it changed, and HOW it was changed.
 Prevent conflicting changes to the same system component.
 Browse and manipulate all components relating to an
application from a single screen, saving you time and
ensuring that changes are complete.
 Automate creation of executables.
 Ensure that the source, executable, and any other form of an
element (module) correspond.

11 Nov, 2003 Introduction to Mainframe 110

 ENDEVOR (Contd.)

 Apply consistent procedures (including the

automation of compiles, impact analysis, and
standards-checking functions) to any component
type, dramatically simplifying the process.
 Process changes to packages and approvals on-
line, eliminating change-related paperwork.
 View or retrieve any prior level of any element.
Report on element definition, content, and
change history.

11 Nov, 2003 Introduction to Mainframe 111

 Summary Slide

 Classification of Computers
 Mainframe Overview
 MVS
 Characteristics Of Mainframe
 MVS – Management of User Jobs
 MVS Concepts
 ISPF – An Introduction
 IBM Utilities – An introduction
 Advanced Mainframe Concepts / Products

11 Nov, 2003 Introduction to Mainframe 112

 THANKS

THANK YOU !

11 Nov, 2003 Introduction to Mainframe 113