Antwort: Re: Weights and CPs
>From my experience I'd avoid to define a uni-processor LPAR. Take at least 2 processors for a LPAR, even the ratio goes over 2.5. Werner Kuehnel IBM Mainframe Discussion List schrieb am 03.02.2010 18:59:00: > Yes, it is not a good idea to assign all of your CPs (physical > processors) to all of your LPARs. What you are doing is having all the > LPARs competing for the CPs on an equal basis. The weights take effect > only when the total CEC is 100% busy or at least very close to it. > Also, the recommendation is that the number of logical processors (LPs) > not be more than 2.5 times the number of CPs. Even in your original > configuration you have have a ratio of 11 LPs to 3 CPs. That is quite > high. > Tom Kelman -- 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
Antwort: Re: Weights and CPs
We also have very small LPARs with 42 and 11 MSUs running on a z10 with 16 procs and turned off HIPERDISPATCH, because it didn't us good. Mainly we saw an increase in IMS resp times when it was turned on. Werner Kuehnel IBM Mainframe Discussion List schrieb am 03.02.2010 19:20:17: > HIPERDISPATCH helps to eliminate this problem. So does IRD VARY CPU > management, but HIPERDISPATCH is much more responsive. HIPERDISPATCH > can park the CPU, which has the same benefit to PR/SM as being offline. > Mark Zelden -- 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
Re: Antwort: Re: Weights and CPs
On Thu, 4 Feb 2010 08:49:23 +0100, Werner Kuehnel wrote: >We also have very small LPARs with 42 and 11 MSUs running on a z10 with 16 >procs and turned off HIPERDISPATCH, because it didn't us good. Mainly we >saw an increase in IMS resp times when it was turned on. >Werner Kuehnel > >IBM Mainframe Discussion List schrieb am 03.02.2010 >19:20:17: > >> HIPERDISPATCH helps to eliminate this problem. So does IRD VARY CPU >> management, but HIPERDISPATCH is much more responsive. HIPERDISPATCH >> can park the CPU, which has the same benefit to PR/SM as being offline. >> Mark Zelden > You turned off HIPERDISPATCH on all LPARs, or just the small ones? I assume you have some bigger LPARs than 42 and 11 MSUs with a 16 engine z10. :-) Mark -- Mark Zelden Sr. Software and Systems Architect - z/OS Team Lead Zurich North America / Farmers Insurance Group - ZFUS G-ITO mailto:mark.zel...@zurichna.com z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/ Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html -- 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