Re: [AMBER] Processors, cores and threads

From: M. Shahid <mohammad.shahid.gmail.com>
Date: Wed, 13 Apr 2011 11:04:32 +0200

Hi,

>From my experience I noticed some speed improvements using HT,
e.g sander minimization of my complex with -np 2 took around 40 min while
the same took 30 min with -np 4.
But as Jason mentioned, I was also "disturbing" the system at the same
time...



-- 
O      o O         o  O    O M.Shahid (MS)
|  o O | | | O o O | | | O  |  Phone: +49-2241-142777
O      O o         O  o    O www.b-it-center.de
On Wed, Apr 13, 2011 at 10:29 AM, Dmitry Mukha <dvmukha.gmail.com> wrote:
> >From my own experience with Intel HT-enabled chips I would recommend not
> to
> take into account HT technology. I did not get any speedup at all comparing
> to -np corresponding to physical core number. Furthermore, performance
> decreased.
>
> 2011/4/12 Hopkins, Robert <hopkins.uhcl.edu>
>
> > Thanks Jason for your background on this issue. It looks like I really
> > have to experiment with -np 2, 4 etc.
> >
> > Bob Hopkins
> >
> >
> > -----Original Message-----
> > From: Jason Swails [mailto:jason.swails.gmail.com]
> > Sent: Tuesday, April 12, 2011 12:00 PM
> > To: AMBER Mailing List
> > Subject: [Possible Spam: 10%] Re: [AMBER] Processors, cores and threads
> >
> > Hi Bob,
> >
> > You can actually run as many threads as you want, the hyperthreading
> > label
> > just says that the cores can handle 2 threads fairly efficiently.  How
> > efficiently, I'm not sure.
> >
> >
>
> --
> Sincerely,
> Dmitry Mukha
> Institute of Bioorganic Chemistry, NAS, Minsk, Belarus
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 13 2011 - 02:30:04 PDT
Custom Search