<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1400" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff background="">
<DIV><FONT face=Arial>Dear David,<BR> only today I've finished
my benchmarks. A little clarification:<BR>i've 2 dual xeon (SMP) 2,66 GHz whith
1GB ram running linux FC2 on a gigabit ethernet. <BR>I've used mpich 1.2.5.2 and
gromacs 3.2.1.<BR>these are the commands i've used:<BR>grompp -f -c -p -po
-shuffle -sort -np # -o<BR>mpirun -np #
/home/Application/gromacs3.2.1/i686-pc-linux-gnu/bin/mdrun -s -x -c -g -e -np #
>& log.log<BR>(#=2,4)<BR>On the dppc system i've obtained this
results:<BR>
(Mnbf/s) (MFlops)
(ps/NODE hour)
(NODE
hour/ns)</FONT></DIV>
<DIV><FONT face=Arial>1 CPU
20.015 721.825
6.184 161.710<BR>2
CPUs
20.747
747.955 6.410 156.000<BR>4
CPUs
34.735 1.252
10.730
93.194<BR> <BR>For the compilation of mpich i've not used any flag. For
fftw and gromacs <BR>I've used the howto indications
(http://www.gromacs.org/documentation/howtos/mpich_howto.html).<BR>So my
question was...do I need to recompile mpich with some flags?<BR>Thanks to
everyone will help<BR>Andrea<BR><BR>----- Original Message ----- <BR>From:
"David" <spoel@xray.bmc.uu.se><BR>To: "Discussion list for GROMACS users"
<gmx-users@gromacs.org><BR>Sent: Friday, June 18, 2004 5:08 PM<BR>Subject:
Re: [gmx-users] Dppc Benchmark<BR><BR><BR>> On Fri, 2004-06-18 at 16:43,
Andrea Carotti wrote:<BR>> > I've run:<BR>> > grompp -f -c -p
-po -shuffle -sort -np 4 -o<BR>> > and then<BR>> > mpirun -np
4<BR>> > /home/Application/gromacs3.2.1/i686-pc-linux-gnu/bin/mdrun -s -x
-c<BR>> > conf_wat.gro -g -e -np 4 >& log.log &<BR>>
<BR>> and what does the performance look like for 1,2,4 processors?<BR>>
Are they SMP or single CPU nodes?<BR>> > <BR>> > ----- Original
Message ----- <BR>> > From: "David" <spoel@xray.bmc.uu.se><BR>>
> To: "Discussion list for GROMACS users"
<gmx-users@gromacs.org><BR>> > Sent: Friday, June 18, 2004 4:35
PM<BR>> > Subject: Re: [gmx-users] Dppc Benchmark<BR>> > <BR>>
> <BR>> > > On Fri, 2004-06-18 at 13:22, Andrea Carotti
wrote:<BR>> > > > Hi all.<BR>> > >
> I' ve the same problem: same hardware and bad
performance..but I've<BR>> > > > MPICH 1.2.5.2, Fedora Core2,
gromacs 3.2.1.<BR>> > > > (I've used the howto:<BR>> > >
> http://www.gromacs.org/documentation/howtos/mpich_howto.html)<BR>> >
> > There is a solution also for me? ...like a flag in the compilation
of<BR>> > mpich<BR>> > > > that i've missed....<BR>> >
> same question then, have you used grompp -shuffle ?<BR>> > > >
TNX in advance<BR>> > > > Andrea<BR>> > > ><BR>> >
> ><BR>> > > > ----- Original Message ----- <BR>> > >
> From: "David" <spoel@xray.bmc.uu.se><BR>> > > > To:
"Seonah Kim" <kim@qtp.ufl.edu>; "Discussion list for GROMACS
users"<BR>> > > > <gmx-users@gromacs.org><BR>> > >
> Sent: Friday, June 18, 2004 10:43 AM<BR>> > > > Subject: Re:
[gmx-users] Dppc Benchmark<BR>> > > ><BR>> > > ><BR>>
> > > > On Thu, 2004-06-17 at 23:23, Seonah Kim wrote:<BR>> >
> > > > Dear all,<BR>> > > > > ><BR>> > >
> > > We are testing our linux cluster now, using gromacs 3.2.1, with
the<BR>> > dppc<BR>> > > > > >
benchmark.<BR>> > > > > ><BR>> > > > > > We
have dual Xeon 2.88 Ghz, 1 GB memory per box. We run RedHat 9.0,<BR>> >
and<BR>> > > > > > use LAM. 512 Mb L2 cache as far as I can
tell.<BR>> > > > > ><BR>> > > > > > We see
135 ps/day for a single processor, but only 250 for two.<BR>> > > >
> > This is not the superscaling reported in the gromacs web site.<BR>>
> > > > ><BR>> > > > > > The two processor run
really works in the same box, using the 2<BR>> > procs,<BR>> > >
> > > we checked, so it is not the network. We read/write only at
the<BR>> > > > > > beggining and end, so it is not the
disk.<BR>> > > > > ><BR>> > > > > > Can
anyone help us ? Is this a LAM issue ?<BR>> > > > > You may need
to recompile using some flags for LAM (short message size<BR>> > > >
> or something like that and set it to 512 kb).<BR>> > > >
><BR>> > > > > Did you run grompp -shuffle?<BR>> > >
> > ><BR>> > > > > > Thanks !!!<BR>> > >
> > > a.<BR>> > > > > ><BR>> > > > >
> _______________________________________________<BR>> > > > >
> gmx-users mailing list<BR>> > > > > >
gmx-users@gromacs.org<BR>> > > > > >
http://www.gromacs.org/mailman/listinfo/gmx-users<BR>> > > > >
> Please don't post (un)subscribe requests to the list. Use the<BR>> >
> > > > www interface or send it to
gmx-users-request@gromacs.org.<BR>> > > > > -- <BR>> > >
> > David.<BR>> > > > ><BR>> >
________________________________________________________________________<BR>>
> > > > David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics
group,<BR>> > > > > Dept. of Cell and Molecular Biology, Uppsala
University.<BR>> > > > > Husargatan 3, Box 596, 75124
Uppsala, Sweden<BR>> > > > > phone: 46 18 471 4205 fax: 46 18 511
755<BR>> > > > > spoel@xray.bmc.uu.se
spoel@gromacs.org http://xray.bmc.uu.se/~spoel<BR>> > >
> ><BR>> >
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<BR>>
> > > ><BR>> > > > >
_______________________________________________<BR>> > > > >
gmx-users mailing list<BR>> > > > > gmx-users@gromacs.org<BR>>
> > > > http://www.gromacs.org/mailman/listinfo/gmx-users<BR>>
> > > > Please don't post (un)subscribe requests to the list. Use
the<BR>> > > > > www interface or send it to
gmx-users-request@gromacs.org.<BR>> > > > ><BR>> > >
><BR>> > > ><BR>> > > >
_______________________________________________<BR>> > > > gmx-users
mailing list<BR>> > > > gmx-users@gromacs.org<BR>> > > >
http://www.gromacs.org/mailman/listinfo/gmx-users<BR>> > > > Please
don't post (un)subscribe requests to the list. Use the<BR>> > > >
www interface or send it to gmx-users-request@gromacs.org.<BR>> > > --
<BR>> > > David.<BR>> > >
________________________________________________________________________<BR>>
> > David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics
group,<BR>> > > Dept. of Cell and Molecular Biology, Uppsala
University.<BR>> > > Husargatan 3, Box 596, 75124 Uppsala,
Sweden<BR>> > > phone: 46 18 471 4205 fax: 46 18 511 755<BR>> >
> spoel@xray.bmc.uu.se spoel@gromacs.org
http://xray.bmc.uu.se/~spoel<BR>> > >
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<BR>>
> ><BR>> > >
_______________________________________________<BR>> > > gmx-users
mailing list<BR>> > > gmx-users@gromacs.org<BR>> > >
http://www.gromacs.org/mailman/listinfo/gmx-users<BR>> > > Please don't
post (un)subscribe requests to the list. Use the<BR>> > > www interface
or send it to gmx-users-request@gromacs.org.<BR>> > <BR>> > <BR>>
> _______________________________________________<BR>> > gmx-users
mailing list<BR>> > gmx-users@gromacs.org<BR>> >
http://www.gromacs.org/mailman/listinfo/gmx-users<BR>> > Please don't post
(un)subscribe requests to the list. Use the <BR>> > www interface or send
it to gmx-users-request@gromacs.org.<BR>> -- <BR>> David.<BR>>
________________________________________________________________________<BR>>
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,<BR>>
Dept. of Cell and Molecular Biology, Uppsala University.<BR>> Husargatan 3,
Box 596, 75124 Uppsala, Sweden<BR>> phone: 46 18 471 4205 fax: 46 18
511 755<BR>> spoel@xray.bmc.uu.se spoel@gromacs.org
http://xray.bmc.uu.se/~spoel<BR>>
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<BR>>
<BR>> _______________________________________________<BR>> gmx-users
mailing list<BR>> gmx-users@gromacs.org<BR>>
http://www.gromacs.org/mailman/listinfo/gmx-users<BR>> Please don't post
(un)subscribe requests to the list. Use the <BR>> www interface or send it to
gmx-users-request@gromacs.org.<BR>> </FONT></DIV></BODY></HTML>