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

Tunning cics region


IBM Mainframe Forums -> CICS
Post new topic   Reply to topic
View previous topic :: View next topic  
Author Message
vicky10001
Warnings : 1

Active User


Joined: 13 Jul 2005
Posts: 136

PostPosted: Thu Aug 21, 2008 11:35 am
Reply with quote

I have created strobe report for tunning cics region. Please find the strobe report

MODULE NAME SECTION NAME PROCEDURE NAME

DB2 DSNGEDM DSNGEPLC LOC SKPT SEC IN DASD/EDM - 11%
NUCLEUS IEATTUSD SUPERVISOR SERVICES - 4%
LELIB CEECCICS CEEVGTSI GET A STACK INCREMENT 3%
NUCLEUS IEAVELK SPIN LOCK SERVICE - 2%
DB2 DSNBBM DSNB1GET RETRIEVE REQUESTED PAGE - 2%
DB2 DSNXGRDS DSNXECW RDS AGENT LOCL NOTIFICTN - 2%

Please let me know any idea for this tunning .. is ther any possible way to tune.. Please give me some suggestion on this
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: Thu Aug 21, 2008 8:13 pm
Reply with quote

Hello,

Which problem(s) are you trying to address?

Why did you run Strobe in the first place?
Back to top
View user's profile Send private message
vicky10001
Warnings : 1

Active User


Joined: 13 Jul 2005
Posts: 136

PostPosted: Fri Aug 22, 2008 10:30 am
Reply with quote

The job is taking more time so we have started tuning cics db2 vsam cobol.
The strobe report its pointing below places

DB2 DSNGEDM DSNGEPLC LOC SKPT SEC IN DASD/EDM - 11%
NUCLEUS IEATTUSD SUPERVISOR SERVICES - 4%
LELIB CEECCICS CEEVGTSI GET A STACK INCREMENT 3%
NUCLEUS IEAVELK SPIN LOCK SERVICE - 2%
DB2 DSNBBM DSNB1GET RETRIEVE REQUESTED PAGE - 2%
DB2 DSNXGRDS DSNXECW RDS AGENT LOCL NOTIFICTN - 2%

So . I need reduce CPU consume time .. Please let me know any idea .

For example
The below strobe report ,CPU consuming is 11% its pointing in DB2.
DB2 DSNGEDM DSNGEPLC LOC SKPT SEC IN DASD/EDM - 11%
Please let me know Any idea to tune this DB2 tunning or Do we need to use any parameter in Db2
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: Sat Aug 23, 2008 3:02 am
Reply with quote

Hello,

Quote:
The job is taking more time so we have started tuning
More time than what? How was the determination that this is "too long"?

If this is cics, which "job"(s) are taking too long?

Are you referring to "wall time"/response time?

One "problem" with looking at database cpu time is that when more cpu% is used it is often because the data was already in the buffer and did not need to be physically read from disk (which would make the run even slower, but may reduce the cpu%).

Have you checked with your cics support (if this is actually a cics issue) about how much load is on the cics region(s)?
Back to top
View user's profile Send private message
vicky10001
Warnings : 1

Active User


Joined: 13 Jul 2005
Posts: 136

PostPosted: Mon Aug 25, 2008 11:24 am
Reply with quote

Thanks for the Help!! Please find the my response

More time than what? How was the determination that this is "too long"?
We are finding High MIps consuming jobs(we are analysing only TOp 50 Jobs) with help of SAS program.

this is cics, which "job"(s) are taking too long?
This region job

Are you referring to "wall time"/response time?
I don't know How to do check "wall time"/response time

My strobe report is pointing the below

DB2 DSNGEDM DSNGEPLC LOC SKPT SEC IN DASD/EDM - 11%

so I need to know DSNGEDM( what is this) .. Why strobe is pointing..
then Is there any possible way to tune this job?
Back to top
View user's profile Send private message
enrico-sorichetti

Superior Member


Joined: 14 Mar 2007
Posts: 10873
Location: italy

PostPosted: Mon Aug 25, 2008 11:32 am
Reply with quote

Quote:
DB2 DSNGEDM DSNGEPLC LOC SKPT SEC IN DASD/EDM - 11%
so I need to know DSNGEDM( what is this) .. Why strobe is pointing..
then Is there any possible way to tune this job?



what measurement do You have that imply that something must be tuned???
from what You have told until now, nothing points out to performance issues
Back to top
View user's profile Send private message
vicky10001
Warnings : 1

Active User


Joined: 13 Jul 2005
Posts: 136

PostPosted: Mon Aug 25, 2008 5:31 pm
Reply with quote

The above statment i took form Strobe report.. Now I got ..

Its taking more time in EDM pool ..

Can I know " what is EDM pool " and What are the parameter or option there in EDM Pool.. Which option is used for allacoate more buffer space.
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 Aug 25, 2008 9:55 pm
Reply with quote

Hello,

For EDM (environmental descriptor manager) look here:
blogs.ittoolbox.com/print.asp?i=18781

Quote:
this is cics, which "job"(s) are taking too long?
This region job
What is "this region job"?

Quote:
We are finding High MIps consuming jobs(we are analysing only TOp 50 Jobs)
Do you mean jobs or cics transactions within one region?

Quote:
Are you referring to "wall time"/response time?
I don't know How to do check "wall time"/response time
Wall time/response time is the time on the clock on the wall (or a wristwatch). It is the least scientific measurement, but often the most visable - people really notice the wait. . .

As has been mentioned before, nothing you posted indicates a performance problem. . .
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 -> CICS

 


Similar Topics
Topic Forum Replies
No new posts Using API Gateway from CICS program CICS 0
No new posts Calling an Open C library function in... CICS 1
No new posts How to 'Ping' a CICS region in JCL CICS 2
No new posts Parallelization in CICS to reduce res... CICS 4
No new posts How to avoid duplicating a CICS Web S... CICS 0
Search our Forums:

Back to Top