[Scip] Timing issues

Giacomo Nannicini giacomo.n at gmail.com
Sun Jul 14 16:06:25 MEST 2013


That makes sense, I was assuming the only trouble would be due to cache.
Thanks a lot.

I got away owing just a couple of beers, not too bad!

Giacomo

On Sun, Jul 14, 2013 at 9:59 PM, Thorsten Koch <koch at zib.de> wrote:
> Dear Giacomo,
>
> if they system has 4 cpus it has a NUMA memory architecture.
> https://en.wikipedia.org/wiki/Non-Uniform_Memory_Access
>
> It is very difficult to get reliable benchmarking out
> of such a machine. We did some experiments with a similar
> system and it is very hard.
>
> First you have to tie processes to cores.
> But this is not enough. You also have to allocated the
> "local" memory of he particular CPU. This is not easily
> done.
>
>> I would
>> expect the jobs to be scheduled on distinct CPUs, so that they would
>> not be competing for L3 cache. Unless the Linux kernel 2.6 scheduler
>> assigns two jobs to the same CPU even if there are free CPUs
>> available?
>
> Can happen.
>
> Best regards,
> Thorsten
>
>
>


More information about the Scip mailing list