[gmx-developers] Benchmark simulation

Roland Schulz roland at utk.edu
Thu Oct 10 12:47:06 CEST 2013

On Thu, Oct 10, 2013 at 5:17 AM, Matthieu.Dreher at imag.fr <
Matthieu.Dreher at imag.fr> wrote:

> Hi,
> I'm currently doing some benchmarking with Gromacs 4.6 and I struggle
> a bit to get scalability after 512 cores. As model I have a 5 million
> coarse grain simulation which should be enough to get a good
> scalability for several thousands of cores. I followed all the steps
> described here
> http://www.gromacs.org/Documentation/Performance_checklist but with
> Intel compiler 12.1 and Intel MPI 4.0.3. I was wondering if there was
> a benchmark suite somewhere for comparaison? It seems there was some
> discussion about it
> (http://comments.gmane.org/gmane.science.biology.gromacs.devel/4513)
> but I don't know if you actually made the data set available.

No. This hasn't been finished. If you send an email with you hardware (e.g.
type of network and cpu) and your md.log performance output to the
gmx-users list, you should get suggestions of how to improve the scaling.


> Best regards
> Dreher Matthieu
> --
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-developers-request at gromacs.org.

ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
865-241-1537, ORNL PO BOX 2008 MS6309
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20131010/7ce2d1ba/attachment.html>

More information about the gromacs.org_gmx-developers mailing list