Work management on ISERIES – for IBM supplied subsystems.
We are accessing data stored on our ISERIES via COGNOS. I am thinking part of our
performance issues is related to the IBM supplied subsystem QUSRWRK.
It is currently only using base pool and non db page faulting is high for the DDM server job.
Is it appropriate to use a storage pool (in addition to the Base pool OR instead of the base pool) to the subsystem QUSRWRK.
Software/Hardware used:
9408-M25 ISERIES
Software/Hardware used:
9408-M25 ISERIES
ASKED:
November 27, 2012 11:23 PM
UPDATED:
November 28, 2012 12:39 PM
Answer Wiki:
Last Wiki Answer Submitted:
Be the first to answer this question.
All Answer Wiki Contributors:
Be the first to answer this question.
Michael Tidmarsh 
11,400 pts. ,
SMOO43 10 pts.
11,400 pts. ,
SMOO43 10 pts. To see all answers submitted to the Answer Wiki: View Answer History.
Is it appropriate to use a storage pool (in addition to the Base pool OR instead of the base pool) to the subsystem QUSRWRK.
If it’s needed, it should be done. However, it should be done to all active subsystems if you’re going to do it to one. Doing it to one might just make things worse.
The initial fundamental reason to do it should be to enable performance measurements. As long as significant work is being routed into *BASE, it’s very difficult to isolate areas that see bottlenecks related to memory and activity levels.
To get quick background, see the QPFRADJ – Set to Automatically needs a IPL? question thread. Then post back here if addtional info is needed.
Tom