Messages in this thread | | | Date | Fri, 16 Nov 2012 18:49:18 +0100 | From | Ingo Molnar <> | Subject | Re: Benchmark results: "Enhanced NUMA scheduling with adaptive affinity" |
| |
* Mel Gorman <mgorman@suse.de> wrote:
> On Fri, Nov 16, 2012 at 04:56:26PM +0100, Ingo Molnar wrote: > > > > * Mel Gorman <mgorman@suse.de> wrote: > > > > > It is important to know how this was configured. I was running > > > one JVM per node and the JVMs were sized that they should fit > > > in the node. [...] > > > > That is not what I tested: as I described it in the mail I > > tested 32 warehouses: i.e. spanning the whole system. > > > > Good (sortof) [...]
Not just 'sortof' good but it appears it's unconditionally good: meanwhile other testers have reproduced the single-JVM speedup with the latest numa/core code as well, so the speedup is not just on my system.
Please post your kernel .config so I can check why the 4x JVM test does not perform so well on your system. Maybe there's something special to your system.
Thanks,
Ingo
| |