[gmx-users] cpu utilization
spoel at xray.bmc.uu.se
Fri Oct 17 19:55:01 CEST 2003
On Fri, 2003-10-17 at 19:37, Dean Johnson wrote:
> Howdy all,
> I have started running some benchmarks using the latest of all the
> required software (with vanilla MPICH). Its a 16 node Dell dual 1.7 &
> 1.8Ghz xeons with gigE. When I am running Amber I get pretty much full
> cpu utillization when I run two 16 cpu (16x1) jobs and slightly less
> with a single 32 cpu (16x2) job. Running GROMACS, I can only get
> utilization up close to 40% when I try larger jobs. The nodes have
> plenty of memory and the systems are pretty quiet. Is this consistent
> with what people are seeing?
I'd suggest using the DPPC benchmark and run grompp -shuffle.
Furthermore for other benchmarks you have to have a reasonable ratio of
computation/communication, e.g. at least 3000-5000 atoms per processor
to scale. It would be interesting to see some scaling benchmarks using
Amber and Gromacs for the same calculation.
> Also, perhaps a bit of evidence, it has a real hard time cleaning up
> after itself. The output indicating performance and stuff comes *way*
> before the jobs eventually die and some never die. I had to write a
> simple 'killgro' command to nuke everything for good. Any ideas?
MPICH problems? Most gromacs users are using LAM. Once more, since TCP
communication in Linux is slow compared to Shared mem, you want to run
8x2 cpus rather than 16x1.
> gmx-users mailing list
> gmx-users at gromacs.org
> 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 van der Spoel, PhD, Assist. 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
More information about the gromacs.org_gmx-users