IBM Mainframe Forum Index
 
Log In
 
IBM Mainframe Forum Index Mainframe: Search IBM Mainframe Forum: FAQ Register
 

How can I compare two jobs in terms of efficiency?


IBM Mainframe Forums -> JCL & VSAM
Post new topic   Reply to topic
View previous topic :: View next topic  
Author Message
ayansau

New User


Joined: 22 Jun 2007
Posts: 42
Location: Chennai

PostPosted: Tue Jan 22, 2008 5:03 pm
Reply with quote

I have two jobs having different steps, performs same operation. How can I compare the two in terms of time taken and space allocated?
Back to top
View user's profile Send private message
Aaru

Senior Member


Joined: 03 Jul 2007
Posts: 1287
Location: Chennai, India

PostPosted: Tue Jan 22, 2008 5:08 pm
Reply with quote

ayanasu,

Have a look at the SYOUT for the time taken and the space used.
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8797
Location: Welsh Wales

PostPosted: Tue Jan 22, 2008 5:14 pm
Reply with quote

Ask your performance and capacity people which process uses least resources.
Back to top
View user's profile Send private message
diwa_thilak

Active User


Joined: 13 Jul 2006
Posts: 205
Location: At my desk

PostPosted: Wed Jan 23, 2008 6:06 pm
Reply with quote

Hi,

You can check out for any tool, like STROBE (which may be shop specific).

It iwll give the time taken by your job and you can use it for your analysis.
Back to top
View user's profile Send private message
khamarutheen

Active Member


Joined: 23 Aug 2005
Posts: 677
Location: NJ

PostPosted: Wed Jan 23, 2008 7:37 pm
Reply with quote

Aya's,

Why does the two jobs perform same operations??? Why cant it be 1 job??? Is there any specific reason for that?
Back to top
View user's profile Send private message
ayansau

New User


Joined: 22 Jun 2007
Posts: 42
Location: Chennai

PostPosted: Fri Jan 25, 2008 11:08 am
Reply with quote

khamarutheen wrote:
Aya's,

Why does the two jobs perform same operations??? Why cant it be 1 job??? Is there any specific reason for that?


This is what called process improvement.

I have made some changes in the code. and I want to measure how much efficient my code is?

I thing I am clear to u
Back to top
View user's profile Send private message
Aaru

Senior Member


Joined: 03 Jul 2007
Posts: 1287
Location: Chennai, India

PostPosted: Fri Jan 25, 2008 11:13 am
Reply with quote

ayanasu,

Quote:
I have made some changes in the code. and I want to measure how much efficient my code is?


Check the CPU time taken by the jobs when run with the same input in similar env. The job which took lesser time for processing the same input is efficient.
Back to top
View user's profile Send private message
ayansau

New User


Joined: 22 Jun 2007
Posts: 42
Location: Chennai

PostPosted: Fri Jan 25, 2008 12:51 pm
Reply with quote

Aaru wrote:
ayanasu,

Quote:
I have made some changes in the code. and I want to measure how much efficient my code is?


Check the CPU time taken by the jobs when run with the same input in similar env. The job which took lesser time for processing the same input is efficient.


Yeh.. Its taking less cpu time, but how can i compare the volume?
Back to top
View user's profile Send private message
dick scherrer

Moderator Emeritus


Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix

PostPosted: Fri Jan 25, 2008 9:24 pm
Reply with quote

Hello,

Quote:
but how can i compare the volume?
Please clarify - which volume? If i understood, both jobs are performing the same operation icon_confused.gif

When you run a batch job, statistics are shown for cpu usage as will as other system resources used. If you talk with the people tasked with performance measuerment or capacity planning, they may have something in place that will provide what you want.
Back to top
View user's profile Send private message
ayansau

New User


Joined: 22 Jun 2007
Posts: 42
Location: Chennai

PostPosted: Mon Jan 28, 2008 9:44 am
Reply with quote

I have removed some sort step from the total process...

Code:
*     6144K  REQUESTED   - VIRTUAL                                            *
*      384K  BELOW 16MEG - FROM USERS PRIVATE AREA                            *
*      348K  BELOW 16MEG - FROM SYSTEM AREA (TOP OF PRIVATE)                  *
*      384K  BELOW 16MEG - FROM USER SUBPOOLS                                 *
*      196K  ABOVE 16MEG - FROM USER SUBPOOLS                                 *
*      348K  BELOW 16MEG - FROM SYSTEM (LSQA+SWA) SUBPOOLS                    *
*    18380K  ABOVE 16MEG - FROM SYSTEM (LSQA+SWA) SUBPOOLS                    *
*        0K  DATA SPACE STORAGE (HIGH WATER MARK)                             *


This is what I am getting at job log. Can anyone tell me the meaning of the volumes...
Back to top
View user's profile Send private message
dick scherrer

Moderator Emeritus


Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix

PostPosted: Mon Jan 28, 2008 8:30 pm
Reply with quote

Hello,

What you have posted shows memory usage above and below the "16-meg" line.
Back to top
View user's profile Send private message
sameer

New User


Joined: 10 Mar 2005
Posts: 41
Location: United States Of America

PostPosted: Wed Apr 02, 2008 1:47 am
Reply with quote

I don't think performance can be measued based on total time the job consumed or total memory it aquired. If you run same job twice, run time may vary. Its depend on CPU usage during the job run.

Coming to point, Measure whether your job running fast after you do code changes, you just check what lines of code you have commented and whether it simplifies the logic whichout affecting the actual expected result.

if so you have done it.

Hope this clarifies you
Back to top
View user's profile Send private message
dick scherrer

Moderator Emeritus


Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix

PostPosted: Wed Apr 02, 2008 2:08 am
Reply with quote

Hello,

Quote:
Hope this clarifies you
Sorry, but it doesn't.

Quote:
run time may vary. Its depend on CPU usage during the job run.
Yes, run time may vary. It does not typically depend on the cpu used. It is the other way around. For the same amount of sequential input, output, and database activity, the cpu used will remain quite constant while the "wall time" may vary considerably.

Quote:
I don't think performance can be measued based on total time the job consumed or total memory it aquired.
Partly correct. . . Some of the main considerations of performance measurement are cpu time, local i/o, and database activity.
Back to top
View user's profile Send private message
View previous topic :: :: View next topic  
Post new topic   Reply to topic View Bookmarks
All times are GMT + 6 Hours
Forum Index -> JCL & VSAM

 


Similar Topics
Topic Forum Replies
No new posts Compare only first records of the fil... SYNCSORT 7
No new posts How to create a list of SAR jobs with... CA Products 3
No new posts Help in Automating Batch JCL jobs mon... JCL & VSAM 3
No new posts Submit multiple jobs from a library t... JCL & VSAM 14
No new posts Compare two files with a key and writ... SYNCSORT 3
Search our Forums:

Back to Top