6.1.09

MAINFRAME










Mainframe Introduction
In this Mainframe tutorial you will learn about mainframes and its concepts and some of the early mainframes describing the evolution and history of mainframes in detail. History and Evolution of Mainframes, Mainframe Channel, Direct Access Storage Device - DASD and Logical Partition - LPAR
What is Mainframe?
As we all know the term mainframe when said to a layman would give a feeling of a large computer, but the question arise how large it is and what is the specialty of these mainframes. The answer to this question lies in the term "Mainframe" itself. As the name implies mainframe is the main computer with all units processing, communications are built into a frame and hence the name Mainframe, which thereby denotes a very large computer.
Mainframes have huge processing and storage capacity. The development cost involved with mainframes is also very huge and therefore mainframes are manufactured only by very large companies. Mainframe gets its usage and applications in banking and financial sectors where large-scale operation with millions of records per day is involved. Apart from the above there is also enterprise class mainframe integration which has greater features and ability of connecting mainframe data sources to distributed platforms.
Thus, with specific, specialized and powerful features of mainframes like security, high performance, reliability, scalability and manageability are achieved and provided to distributed applications using mainframe resources.
History and Evolution of Mainframes:
Some of the early mainframes which were developed starting from the year of 1942 are ENIAC, MARK1, BINAC, UNIVAC. ENIAC is also called as electronic numerical integrator and calculator was developed in the year 1942. This mainframe machine weighed in tones and consumed enormous electric power.
It had thousands of vacuum tubes, relays resistors, capacitors, and inductors inside it. In the year 1951, UNIVAC-I was developed specially for the US Census Bureau. The major difference between UNIVAC and ENIAC was the processing of digits. In UNIVAC, processing was done in serial mode, yet the speed of UNIVAC was higher than ENIAC with one disadvantage of vacuum tubes generating enormous amount of heat which made the mandatory requirement of a large good air conditioning system.
Later in 1954, Univac II was developed. In 1962, the concept of virtual memory, a powerful feature which shall be discussed in later sections, was introduced in mainframes making the system even more powerful. In 1964, the first third generation computer named as Burroughs B5000 came into market. This mainframe system had various powerful specialized features like multiprogramming, multiprocessing and virtual memory.

Later on various improved versions of Burroughs series came into market with new features in each. Among this the most notable one was B6700 which came in the year 1971 and this supported the feature of dynamic linking of programs at runtime.
IBM was producing and releasing mainframes in the market at all periods from past till present with the successive development of IBM Series starting with System/360. One of the biggest factors associated with mainframes is its huge cost which made its usage only possible for big companies and corporate. There are still many technical terms one must get familiarized with mainframe systems to understand about mainframe systems in depth.
Let us see some of the internal technical usage of mainframe system.
Mainframe Channel:
A mainframe channel connect to one or more controllers via either pairs of large "bus and tag" cables or, fiber optic ESCON (Enterprise System CONnection) cables and FICON, which has the ability controlling one or more devices. This is one of the important term in mainframe technology since it has the ability of take care of huge input and output functions.

DASD:
DASD stands for Direct Access Storage Device. This indicates to any type of storage that was directly (randomly) addressable.

LPAR:
LPAR stands for Logical Partition and is a powerful hardware or firmware feature implemented in all mainframe systems. By this feature it is possible to create partitions and by which CPUs and I/O sub-systems can be shared between logical partitions.
JCL INTRODUCTION

Job control language. It is a means of communication between a program that
can be written in COBOL , ASSEMBER or PL/I and the MVS operating system. Without
a JCL, you cant able to run a job on MVS operating system.

Let us start with an example jcl, how it looks like, The following jcl is
used to run an cobol program. I will explain in this chapter each and every
line of this jcl. If you are already fimiliar with jcl,
JCL , used to run a cobol program

//JOB1 JOB (034D),'RAMESH KRISHNA REDDY',CLASS='A',PRTY=6
//STEP01 EXEC PGM=COBPROG.
//INFILE DD DSN=SED.GLOB.DES.INFILE,DISP=SHR
//OUTIFLE DD DSN=SED.GLOB.DES.OUTFILE,
// DISP=(NEW,CATLG,DELETE),
// UNIT=DISK,
// SPACE=(CYL,(1,5),RLSE),
// DCB=(RECFM=FB,LERECL=70,BLKSIZE=700)

to see source code of cobol program COBPROG click here
Now let us discuss JCL which we have seen.

Basics about JCL
All JCL statements can consists of up to five types of
fields

// The two forward slashes are required at the beginning of each
JCL statement in Columns 1 and 2.

Name field - This is an optional field. If coded, should start at
Column 3. It maximum length is 8.

Operation field - Which indicates the operation that is to be performed

Operand - This field must appear after Operation field. Must start at 16
column

Comments - Comments begin one space after the last operand. Our simple
example has no comments.


// Name Operation OPerand

// JOB1 JOB (034D),
'RAMESH',
CLASS='A',
PRTY=6

// STEP01 EXEC PGM=COBPROG

// INFILE DD DSN=SED.GLOB.DES.INFILE,
DISP=SHR
// OUTIFLE DD DSN=SED.GLOB.DES.OUTFILE,
// DISP=(NEW,CATLG,DELETE),
// UNIT=DISK,
// SPACE=(CYL,(1,5),RLSE),
// DCB=(RECFM=FB,
// LERECL=70,BLKSIZE=700)

Three important input operations are
1. JOB
2. EXEC
3. DD


1.JOB statement

EXPLANATION to keyword parameters
1.CLASS
2.PRTY
3.MSGCLASS
4.MSGLEVEL
5.TYPRUN
6.NOTIFY

1. CLASS - Some jobs can be short running, some jobs may take more time other
may use havy resources. CLASS keyword parameter is used to tell to
OS about the nature of job we are submitting.

Syntax -> CLASS=jobclass

Valid values -> Any alpha betical character between A - Z or
numbers between 0 - 9

Example JCL -> //MYJOB JOB (3345Y),'KRISHNA REDDY',CLASS=K

2. PRTY - It is related to CLASS parameter. It assigns priority to jobs which
belong to the same class. Higher number takes precedence over the
job with lower number. A job with priority of 12 will run before a
job with priority 5.

Syntax -> PRTY=priority

Valid values -> Any value between 0 to 15

Example JCL -> //MYJOB JOB (345Y),'KRISHNA REDDY',CLASS=8,PRTY=6

3. MSGCLASS - The MSGCLASS parameter determines the output device to which
system messages and JCL messages are written

Syntax -> MSGCLASS=output-class-name

Valid values -> Any alpha betical character between A - Z or
number between 0 - 9

Example JCL -> //MYJOB JOB (456Y),'KRISHNA REDDY',CLASS=7,PRTY=5,MSGCLASS=A

4. MSGLEVEL - is used to tell JCL, which messages to be printed in the device
specified in MSGCLASS parameter

Syntax -> MSGLEVEL=(statements,messages)

Valid values -> Statements may be the number 0,1,2 and messages can be 0 or 1

STATEMENT
0 - Related to job statements only printed
1 - jcl will be printed
(which includes all cataloged/symbolic parameters)
2 - only the input jcl statements being printed

MESSAGES
0 - Allocation/termination messages will be printed
If job terminates abnormally
1 - Allocation/termination messages will be printed
regardless of whether the job terminates normally or
abnormally.

Example JCL -> //MYJOB JOB (3456Y),'SR',CLASS=8,MSGCLASS=S,MSGLEVEL=(1,1)

5. TYPRUN - The TYPRUN parameter is used to specify whether job is to be held
until further notice or for syntax checking

Syntax -> TYPRUN=HOLD - Job held until further notice
or
TYPRUN=SCAN - Check Jcl for syntax checking

Example JCL -> //MYJOB JOB (3456Y),'KRISHNA',CLASS=8,PRTY=9,TYPRUN=HOLD


6. NOTIFY - The NOTIFY parameter is used to direct the system, where it has to
send the success/failure message after completing the job.

Syntax -> NOTIFY=userid/&SYSUID

EXAMPLE JCL -> //MYJOB JOB (3456Y),'KRISHNA REDDY',CLASS=8,NOTIFY=&SYSUID

&SYSUID - the userid from which user it has been submited

//MYJOB JOB (34W5),'KRISHNA REDDY',CLASS=8,NOTIFY=ERT54
It send the sucess/failture message to ERT54 userid

2.EXEC statement

->EXEC statement is used to execute a program/procedure

->A maximum of 255 EXEC statement can code in an single job

Syntax - //stepname EXEC PGM=program-name,keyword parameters
Positional parameter - Program-name

Keyword parameters for EXEC
1.PARM
2.ACCT
3.ADDRSPC
4.DPRTY
5.PERFORM
6.RD

PARM
PARAM parameter is used to pass information to program
Syntax -> PARM=value
Value is a string can be 1 to 100 characters long

PASSING PARAMETER USING PARM PARAMETER
ACCT - accounting information for that step

ADDRSPC - used to indicate to the system that the job step is use either
virtual or real storage

DPRTY - used to assign priority to the job step

PERFORM - specifies the rate at which system resources used by job step

RD - restart definition is used to specify automatic restart of a job
if it abends

NOTE
Q. If there is a situation, where we need to code more than 255 steps in a JOB?
A. We need to split jcl into two jcls , at the end of the first jcl check the conditioncode and initiate the second jcl.


PARAMETERS IN BOTH STATEMENTS (JOB and EXEC)
1.COND
2.REGION
3.TIME

These parameters can code in both JOB and EXEC statements, When we code
these parameters on both statement,

JOB - REGION will override EXEC - REGION

JOB - COND will override EXEC - COND

EXEC - TIME will overrirde JOB - TIME


1.REGION - Specifies the amount of space should be used at the time of
running a job/step

Syntax -> REGION=<>K
or
REGION=<>M

Example JCL -> //MYJOB JOB (AE32),'RAMESH'
//STEP1 EXEC PGM=COBPROG,
// REGION=67K

NOTE:All available storage is assigned to the job or jobstep, If REGION is coded 0K or 0M

2.COND - Each step sends an a return code to the system upon completion.
This is called condition code. COND parameter is used to control
the execution of subsequent job steps, depending on the condition
code that is returned for prior step.

Return code number between 0 and 4095

Syntax -> COND=(comparsion-code,condition)
(If coded on job)

COND=(comparsion-code,condition [,stepname] [,EVEN / ONLY])
(If coded on step)

condition can be GT,GE,LT,LE,EQ,NE

Example JCL -> COND coded on JOB

//MYJOB JOB (R475),'KRISHNA REDDY'
// COND=(4,GT)
//STEP1 EXEC PGM=COBPROG
//STEP2 EXEC PGM=COBPROG2

In this example we are specifing that if the number
4 is grater than the return code of STEP1, then STEP2
is to be bypassed. It is elaborated further in the
following image


COND coded on EXEC statement

//MYJOB JOB (U769),'KRISHNA REDDY'
//STEP1 EXEC PGM=PROG1
//STEP2 EXEC PGM=COBPROG,
// COND=(8,EQ,STEP1)
//

In this example the COND parameter is used to specify that
STEP2 should be by passed if 8 is equal to the return code
issued by STEP1.

TIME - The time parameter is used to specify the amount of CPU time that a
job or job step is permitted to utilize.

Syntax -> TIME=minutes
OR
TIME=([minutes] [,seconds])

minutes can be between 1 and 1439
seconds can be between 1 and 59

Example JCL -> If coded on JOB

//MYJOB JOB (E234),'RAMESH KRISHNA',TIME=(20,30)
//STEP1 EXEC PGM=COBPROG1
//STEP2 EXEC PGM=COBPROG2
//STEP3 EXEC PGM=COBPROG3

In this example 20 minutes 30 seconds time alloted
to job. All steps in this job STEP1 STEP2 STEP3 should
complete its task within 20 min. 30 sec.

If coded on STEP

//MYJOB JOB (R567),'KRISHNA'
//STEP1 EXEC PGM=COBPRO
// TIME=30

In this example 30 min. time is alloted to STEP1.
If STEP1 requires more than 30 min. MYJOB will
terminate abnormally.

If coded on both STEP AND JOB

//SECOND JOB ,'R. KRISHNA',TIME=3
//STEP1 EXEC PGM=C,TIME=2
//STEP2 EXEC PGM=D,TIME=2


In this example, the job is allowed 3 minutes of execution time.
Each step is allowed 2 minutes of execution time. Should either
step try to execute beyond 2 minutes, the job will terminate
beginning with that step. If STEP1 executes in 1.74 minutes and
if STEP2 tries to execute beyond 1.26 minutes, the job will be
terminated because of the 3-minute time limit specified on the
JOB statement.

By coding TIME=1440 or TIME=NOLIMIT, It will give a job or step an unlimited amount of time.

To allow a job or step to use the maximum amount of time, code TIME=MAXIMUM. Coding TIME=MAXIMUM allows the job or step to run for 357912 minutes. (Approx. 248 days)

3.DD STATEMENT
The DD statement (Data Definition), is used to identify the source of input
and the placement of output information

Syntax -> //ddname DD <>

ddname must be unique in the job

Positional parameters - *
DATA
DUMMY
DYNAM

Keyword Parameters - DSN
DISP
UNIT
SPACE
DCB
VOLUME

DSN PARAMETER

DSN parameter is used specify the data set name

Syntax -> DSN=dataset name

Example JCL -> //MYJOB JOB (ER3),'RAMESH R'
//STEP1 EXEC PGM=COBPROG
//INFILE DD DSN=TEST.GLOB.MYLIB

TEST.GLOB.MYLIB will be used in the COBPROG program.

TEMPORARY DATA SETS

Temporary data set is created during job and deleted at the end of the job.

Temporary data set can be coded by using two ampersands followed by name.

Example JCL --> //MYJOB JOB (E456),'RAMESH'
//STEP1 EXEC PGM=COBPROG
//INFILE DD DSN=&&TEMP

Here TEMP is an temporary dataset will be deleted upon job completion.

DISP PARAMETER
The DISP parameter is used specify the disposition of dataset which is
coded on DSN parameter.

Syntax ->

DISP=(STATUS,NORMAL,ABNORMAL)

Parameter on the DISP statement


Status Normal Disposition Abnormal Disposition
NEW DELETE DELETE
OLD CATLG CATLG
MOD UNCATLG KEEP
SHR KEEP UNCATLG
PASS

STATUS

NEW - Dataset will be created. (file should not exists)
OLD - Dataset should exists.
MOD - Dataset will be created If not exists.
SHR - Dataset can be used by other jobs also

NORMAL DISPOSITION
(Happened upon sucessful execution of job step)

DELETE - Dataset should be deleted
CATLG - Dataset will be cataloged
UNCATLG - Dataset will be removed from system catalogs
KEEP - Dataset will be retained (This parameter should
be used with permanent data sets)
PASS - Dataset is to be passed subsequent job step in
the same job



ABNORMAL DISPOSITION
(Happened upon unsucessful execution of job step)

DELETE - Dataset should be deleted
CATLG - Dataset will be cataloged
UNCATLG - Dataset will be removed from system catalogs
KEEP - Dataset is to be kept

EXAMPLE JCL --> //MYJOB JOB (E674),'KRISHNA REDDY'
//STEP1 EXEC PGM=COBPROG
//INFILE DD DSN=TEST.GLOB.TEMP,
// DISP=(NEW,CATLG,DELETE)


In this example ,

DISP=(NEW,CATLG,DELETE)
NEW - TEST.GLOB.TEMP is not exists, it will be created
CATLG - Upon successful execution of job step, data set
will be cataloged
DELETE - If job terminicated abnormally, dataset will be deleted

UNIT PARAMETER
In IBM Mainframe environment, All devices have an address assigned to
them at the time they added to the sytem. Devices can be referenced
using this addresses. UNIT parameter is used to specify thise address.

Syntax -> UNIT=device_address/device_type/device_group_name/TAPE

EXAMPLE JCL -> //MYJOB JOB (R345),'KRISHNA REDDY'
//STEP1 EXEC PGM=COBPROG
//INFILE DD DSN=TEST.GLOB.TEST.LIB,
// UNIT=SYSDA

In above example, COBPROG is executed, and a file INFILE which may
reside on any of the devices which are grouped under the symbolic name
SYSDA will be accessed

VOL PARAMETER

This parameter is used to identify the volume serial number on which
dataset is reside. The VOL dataset is used with disk and tape datasets.

Syntax -> VOL= volumelabel/data set label

Sub parameters used with VOL parameter

SER - Specification of serial number
REF - Referencing VOL specification from a prior step
PRIVATE - Allowing access to volume by single user
RETAIN - Inhibiting dismounting of volume until end of job
SEQ - Specification of sequence in which volumes are to be mounted

EXAMPLE JCL --> //MYJOB JOB (E454),'KRISHNA REDDY'
//STEP1 EXEC PGM=COBPROG
//INFILE DD DSN=TEST.GLOB.TEMP,
// VOL=SER=(VOL1,VOL2,VOL3)

In this example the data set called DATA3 resides on 3 volumes whose
serial numbers are VOL1,VOL2,VOL3. The Operating system will request
that all volumes be mounted at the same time.

SPACE PARAMETER

The SPACE parameter is used to allocate space for datasets.
We can allocate space in Cylinders/Tracks/Blocks

Syntax -> SPACE=(CYL,(primary,secondary,directory),RLSE,CONTIG,MXIG,ROUND)

Instead of CYL, We can use TRK or BLK


Meaning of Sub Parameter

TRK - Requesting space in track

CYL - Requesting space in cylinders

PRIMARY - Primary storage to be allocated at the time of
data set created

SECONDARY - Additional storage to be allocated ,
If primary storage is not sufficient

DIRECTORY - Space for recording of name and location of partitioned
data sets

RLSE - Request for release of space previously allocated unused
space after completion of job

CONTIG - Request for contiguous space

MXIG - Request for large aread of contiguous space

ROUND - Request for entire cylinder for storage of data set


EXAMPLE JCL -> //MYJOB JOB (W345),'KRISHNA REDDY'
//STEP1 EXEC PGM=COBPROG
//INFILE DD DSN=TEST.GLOB.LIB
// UNIT=4560
// SPACE=(CYL,(30,4))

In this example, 30 cylinders are requested as primary space and 4
additional cyliders as secondary space.

Upto 15 extends of secondary space ( in our example it is 4 ) will be alloted , If space is not sufficient.


DCB PARAMETER

Records in dataset may contain FIXED length / VARIABLE length.
The DCB (Data Control Block) parameter is used to specify record
format,record length, block size etc..

Syntax -> //ddname DD DCB=<>

Subparameters in DCB

RECFM - Specification of record format - F/FB/V/VB/U
LRECL - Specification of record length
BLKSIZE - Specification of block size
BUFNO - Specification of buffers


EXAMPLE JCL -> //MYJCL JOB (E3445),'RAMESH'
//STEP1 EXEC PROG=COBPROG
//INFILE DD DSN=TEST.GLOB.LIB
// UNIT=234,
// DSN=(LRECL=80,
// RECFM=FB,
// BLKSIZE=800,
// BUFNO=30)


In this example, The DCB parameter specifies that this file is to
have a logical record length of 80 bytes, it will have a fixed block
record format, and the block will 800 (800 is multiple of 80). The
BUFNO parameter is set to 30, indicating upto 30 buffers may be
utilized in virtual storage for this data set.

Default buffers are 5,if you not specified any thing

When you specified V for RECFM parameter, LRECL value is largest record in the file plus 4 bytes. These four bytes contain the actual length of each variable length record in the file

No comments:

Post a Comment