>I would say the majority of shops I went into way overused IMP=1 and >at the same time underused IMP=5.
>But from my experience at a lot of shops, >I think many of them just used IMP=1 for just about everything considered >"production online". All CICS regions, all DB2 subsystems, MQ, etc. Then >they would (misguidedly) try and use velocity to make some distinction >within that service class. Have you been to our shop here, too, Mark? And did not come by my desk?!? <vbg> When I first defined the WLM policy here, I had it all spread out evenly. And then (since everyone knew we went to WLM mode) everyone looked at the system and started screaming once 'their' product had delays. Never mind that those delays would also have been there in compat mode! Never mind that the PI was still good. In the end I was forced to classify just about everyone (in our case IMS and DB2) in an IMP1 service class. Those that screamed loudest got promoted to imp1. We have just about nothing anymore in imp4 and 5 (but a lot in discretionary - I made sure of that at least). >your only >option will be to decrease the importance of all other IMP1 address spaces. So true. We are there. Same caveat as before: How dare we (my two colleagues and I) even suggest to take 'their' product out of an imp1 class? How dare we suggest that not all STCs are created equal or have equal priority? And have you noticed how many 'new' products make themselves go into at least SYSSTC by 'absolutely needing' a SCHEDxx entry? When I rewrote the policy (still haven't been allowed to do rollout), I classified them all ahead of the SPM rule for SYSSTC, of course at a lower importance! Regards, Barbara Nitz ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
