Le Jeudi 12 Septembre 2002 02:34, DeLano, Warren a écrit :
> Has anyone out there run any performance comparisons between single and
> dual processor systems based on the AMD Athlon MP?
>
> If one is running two independent jobs (not parallel, just two sanders),
> does one really get full performance on each job? or is there contention
> over the RAM or CPU buses which would result in less than full CPU
> utilization? If so, then what's the magnitude of this penalty?
>
> Thanks,
> Warren
If you use a dual-athlon for one job, you'll get nearly the same performance
as two one-node athlons. So a dual athlon is the best best. Meanwhile amber
suffers from lack of scaling in this configuration.
If you use each processor of the dual node separately, you get roughly what
you want (one hundred percent of one processor for each job).
Stef
--
*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*
Teletchéa Stéphane - CNRS UMR 8601
Lab. de chimie et biochimie pharmacologiques et toxicologiques
45 rue des Saints-Peres 75270 Paris cedex 06
tel : (33) - 1 42 86 20 86 - fax : (33) - 1 42 86 83 87
mél : steletch_at_biomedicale.univ-paris5.fr
*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*
Received on Thu Sep 12 2002 - 01:09:58 PDT
