On Fri, 2006-09-01 at 12:37 -0400, Veilleux, Jon L wrote: > I would like to see a pOOP (Performance monitoring Offload Optimizer) > for running RMF, OMEGAMON, etc. That would save a TON of CPU cycles. The > performance monitoring suite seems to take more cycles than all of the > applications they are monitoring.
Is this really still an issue Jon ???. Monitoring options of course can affect this markedly - especially things like traces being left on. Since RMF offered services that do a lot of the grunt work seems the vendors re-wrote a lot of their stuff to use that, and cut down the SUs their address spaces were charged. Don't know about Omegamon, as I haven't looked at their numbers, but certainly TMON dropped quite a bit. I believe the same applies to the R11 (and later presumably) CA offerings. Certainly it used to be the case that the monitors were subject to being terminated on small LPARs if their consumption got too high. Seems less of an issue for me these days - subject to the Omegamon caveat above. Where I get queries is on LPARs where there is predominantly enclave workloads - SAP R3 say. People watching SDSF see a distorted view of the world, and it looks like the monitors are out of whack re CPU consumption. 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
