[gmx-users] gigabit scaling, DPPC benchmark

[David]

On Tue, 2005-06-07 at 14:23 -0400, Jason de Joannis wrote:
> > Date: Tue, 7 Jun 2005 17:32:47 +0200 (CEST)
> > From: David van der Spoel <spoel at xray.bmc.uu.se>
> > Subject: Re: [gmx-users] gigabit scaling, DPPC benchmark
> > To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> > Message-ID:
> >  <Pine.LNX.4.44.0506071732001.17422-100000 at rembrandt.bmc.uu.se>
> > Content-Type: TEXT/PLAIN; charset=US-ASCII
> >
> > On Tue, 7 Jun 2005, Jason de Joannis wrote:
> > >>
> > >> > Then why does the Xeon2800/1000Gbit system on gromacs.org scale so well?
> > >> > I didn't see any special comments about its network.
> > >> Where do you find that? You don't mean the old scaling benchmarks on 800
> > >> MHz PIII machines?
> > >Not exactly. I am talking about the last line of the 3.0 scaling
> > >benchmarks. For example, the 4 processor job runs at 637 ps/day which
> > >is 124% scaling. By contrast my machine which is quite similar as far
> > >as I can tell, gets only 517 ps/day or 81% scaling. Are those benchmarks
> > >for quad-nodes? Mine are dual-nodes.
> > Aha, I wasn't aware of these benchmarks.
> > These are most likely dual nodes.
> >
> > Did you use -shuffle -sort?
> 
> yes I did.
then it could still be hardware: switch, network card. Not all switches
are equal. Maybe you can try other performance tests, if I'm not
mistaken LAM comes with a set of tests.
> 
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,          75124 Uppsala, Sweden
phone:  46 18 471 4205          fax: 46 18 511 755
spoel at xray.bmc.uu.se    spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++