Mark Jacobs wrote:
We have two lpars that perform the vast majority of our batch work. CA7 runs on one of these lpars and most of the CA7 submitted jobs are picked up by the JES2 on that system. This leaves the other system almost empty of batch work while the first system is being over loaded.
That's just how JES2 works. The member holding the checkpoint always attempts to schedule all available work. I've discussed this phenomenon many times.
What are some of our options to spread the workload across both systems on an equal basis.
In spite of it's selfish behavior, which can't be changed, a JES2 member will always stop initiating work when no initiators are left to be populated. Therefore, if you limit the number of available initiators on the overworked system, the other system will pick up the slack.
zOS 1.4. JES2 managed batch. Thruput manager is available.
You might consider the use of WLM-managed batch initiators. Beginning with z/OS V1R4, it more aggressively "trims" the initiator pool on overworked systems.
-- Edward E Jaffe Phoenix Software International, Inc 5200 W Century Blvd, Suite 800 Los Angeles, CA 90045 310-338-0400 x318 [EMAIL PROTECTED] http://www.phoenixsoftware.com/ ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
