The consideration is the LPAR overhead. The more dispatching there is, the more is the "physical overhead". I've seen that running as high as 5% of each processor on overcommitted systems, and .1% normally.
Joe Martin wrote:
Have a development system that has 2 IFLs and both are configured to the LPAR running z/VM. Have several SUSE Linux 11 guests running and the LPAR is busy during the day - high 90s percent busy as displayed by Velocity. I noticed yesterday that two of the busier systems have *3* vCPUs configured to them. I'm wondering if that will hurt or help performance of the guest (or maybe be of no consequence) when the LPAR is very busy. Thanks, Joe ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/
---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/
<<attachment: BARTON.vcf>>