Archives


- Beowulf
- Beowulf Announce
- Scyld-users
- Beowulf on Debian

[Beowulf] [gorelsky@stanford.edu: CCL:dual-coreOpteron275performance]

Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.

Search

Kozin, I (Igor) i.kozin at dl.ac.uk
Thu Jul 14 08:42:00 PDT 2005


taskset is part of schedutils by Robert Love http://rlove.org/schedutils/

> >> But now for 4cores/2CPUs per Opteron node to force the using of
> >> only 2 cores (from 4), by 1 for each chip, we'll need to have
> >> cpu affinity support in Linux.
> >
> >Mikhail,
> >you can use "taskset" for that purpose. 
> >For example, (perhaps not in the most elegant form)
> >	mpiexec  -n 1 taskset -c 0 $code : -n 1 taskset -c 2 $code
> >But I doubt you want to let the idle cores to do something else 
> >in the mean time. However small you will generally see an increase 
> >in performance if you use all the cores.
> 
> Thanks !
> AFAIK, taskset isn't a part of numactl package. Sorry, where is 
> possible to download taskset ?
> 
> What is about using of all the cores, you are, of course, in
> general right ;-)
> But there may be some cases where using of "pairs of cores" on Opteron
> is bad under parallelization. For example, test178/RHF on the same
> G03 w/Linda, gave more worse performance on 2 cores than for 1 core
> (because of extra memory traffic).
> And, again at least theoretically, I may "occupy" 2 (free) of total
> 4 cores in 2-chip server for example w/some independed cache-friendly 
> tasks.
> 
> Yours
> Mikhail  
>    
> 
> Yours
> Mikhail
> 
> 
> 
> >
> >Best,
> >Igor
> >
> >
> >I. Kozin  (i.kozin at dl.ac.uk)
> >CCLRC Daresbury Laboratory
> >tel: 01925 603308
> >http://www.cse.clrc.ac.uk/disco
> >Distributed Computing Forum
> >http://www.cse.clrc.ac.uk/disco/forums/ubbthreads.php?Cat=0
> >
> >_______________________________________________
> >Beowulf mailing list, Beowulf at beowulf.org
> >To change your subscription (digest mode or unsubscribe) visit 
> >http://www.beowulf.org/mailman/listinfo/beowulf
> 
> 




More information about the Beowulf mailing list