>Date: Mon, 15 Oct 2007 10:57:09 +0200 >From: Barbara Nitz <[EMAIL PROTECTED]> >Subject: Re: Are there tasks that don't play by WLM's rules >We run one of those newfangled things called WBIFN (MQS-Merva-Bridge to swiftnet). That productive service class >runs IMP1 exvel 40% (I believe, not that that makes much difference) on a 2 processor lpar dedicated to that >workload and has about 65 address spaces.
Hi Barbara, long time no see. This kind of workload has always been a problem for Velocity goals. You have 2 CPs and 65 address spaces. When very busy you will see 2 A/S running and 63 with CPU delay. The large number of CPU delay samples will bring the velocity down significantly, even though the delay is not large. At quiet times with many of the 65 address spaces in Wait State, the velocity might be fine. I have seen this same effect with a Peoplesoft customer trying to use velocity for the DDF enclaves. Unfortunately you are stuck if your application doesn't support response time goals but has a large swings in MPL. This has been a problem with managing with velocity goals since the beginning. Tom Russell, Internet: [EMAIL PROTECTED] "Stay calm. Be brave. Wait for the signs." -- Jasper FriendlyBear "... and remember to leave good news alone." -- Gracie HeavyHand ---------------------------------------------------------------------- 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
