Home > IT Answers > DataCenter > Performance ‘Blip’
BarbaraS
0 pts.
 Performance ‘Blip’
DataCenter
We have a heavy batch atmosphere, at certain peak times, for only about 8 seconds, it appears the system is doing nothing. Our main subsystem and its subsequent jobs are set up to come out of *base pool with a max act level of 687. There are usually around 900 jobs with 1000 threads running. Should the max act level be raised, or is there a possible problem with journaling?

Software/Hardware used:
ASKED: June 29, 2006  1:21 PM
UPDATED: July 10, 2006  7:43 AM
RELATED QUESTIONS
Answer Wiki:
First get your user subsystems out of *BASE (CHGSBSD)and into their own memory pools. *Base is like a traffic cop and needs to be dedicated to system work as much as possible. Put the user subsystems into shared pools WRKSHRPOOL and set the the QPFRADJ system value to automatically adjust memory pools. If you still see periods where nothing appears to be happening consider more memory or CPU upgrade.
Last Wiki Answer Submitted:  July 10, 2006  7:43 am  by  JDWWms   0 pts.
All Answer Wiki Contributors:  JDWWms   0 pts.
To see all answers submitted to the Answer Wiki: View Answer History.


RSS - Discussions Help
Discuss This Question:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _