>>> On 2/5/2010 at 07:04 PM, Patrick Spinler <[email protected]> wrote: > I'd plan to do all this testing on an up to date install of rhel 5.4 in > a single IFL guest, ...
Comparing the performance of one instance of Linux for System z to one instance of anything else is likely going to make the System z look anemic, or at the best, have equivalent performance at a much higher cost. System z will shine only if there are multiple instances running, and performance doesn't degrade all that much because your guests are set up properly. Second, artificial benchmarks are typically oriented towards midrange system, single instance performance. That includes iozone. You would be much better off getting your hands on something that will drive real workload, repeatably. That way, when you compare 50 Linux guests running against 50 distributed systems, you'll have a reasonable comparison. Mark Post ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
