Thanks Mike, I was thinking that it might be better to consolidate the 2 Prod LPARs so that all of the workload operated under the same WLM policy without the PRSM time slice. But that's not my call.
Thank You, Dave O'Brien NIH Contractor ________________________________________ From: Mike Schwab [[email protected]] Sent: Tuesday, November 30, 2010 1:28 PM To: [email protected] Subject: Re: Question concerning CMF reporting of CPU Delay On Tue, Nov 30, 2010 at 9:35 AM, O'Brien, David W. (NIH/CIT) [C] <[email protected]> wrote: > Hi Kees, > > 3 physical CPs > 2 Prod LPARs each with 3 LPs > 1 Test LPAR with 2 LPs > 1 Sandbox with 1 LP > > So we have 9 LPs competing for 3 CPs. Fortunately 3 of those LPs are little > used. > The relative weights of the Prod LPARs are 45 and 30. Maybe a 42 / 33 weight would work a little bit better? > The LPAR reporting the problem is the one with 30, no surprise there. > The Test LPAR has a weight of 15 while the Sandbox has a weight of 10 and > seldom use their relative share. > > I have suggested that a 2:1 LP to CP ratio would be more efficient but was > told the faster CP make the 2:1 ratio obsolete. Makes no sense to me but then > I'm the DASD guy. Every problem is HSM until proven otherwise. > > Thank You, > Dave O'Brien > NIH Contractor -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- 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 ---------------------------------------------------------------------- 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
