On Fri, 2007-10-12 at 12:04 -0500, Kelman, Tom wrote: > We are about to set the soft cap on our systems. I have been informed > by my system programmers that in the past they have seen tasks that > don't play by WLM's rules when it comes to resource utilization. One is > the Direct-Connect (NDM) task. In the past, before I came to this > location, they saw Direct-Connect take CPU resources and affect CICS > processing when the system was heavily used even though Direct-Connect > had an importance level lower than the CICS regions. We are running NDM > at an importance level of 4 with a velocity goal of 20. Our CICS > regions are at an importance level of 1 with a velocity goal of 50. The > CICS transactions are run at an importance level of 1 with a response > time goal of 80% within 0.5 seconds.
If you are running to transaction goals, your imp/velocity is irrelevant. What I suspect happened is that your CICS got driven back to it's goal. In an unconstrained environment your users probably got used to "good" (i.e. better than the goal) response times. Easy enough to check from RMF. An enforced "80% at half a second" ain't all that flash maybe after being on a good roll. So no, I'd say everyone is behaving as expected - the machine (WLM) is probably just doing what it was told. Short answer, fix the goal(s) - and ensure enough resource is available to meet them. Shane ... ---------------------------------------------------------------------- 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
