[gmx-users] Scaling on IBM/SP2 Power3
David van der Spoel
spoel at xray.bmc.uu.se
Wed Nov 6 00:04:14 CET 2002
On Wed, 2002-11-06 at 12:26, Anton Feenstra wrote:
> > > I'd appreciate any comments and/or other suggestions!
> > You are omitting the most important information: is this with PME or
> > with a cut-off? With cut-off you should be able to reproduce the gmx
> > benchmarks, with PME it won't be as nice... Erik (and me) are working on
> > this though...
> That is one of the reasons I have not (yet) switched to using PME... ;-}
> So, I'm using cut-offs which should give me better scaling than I have.
OK, but since your system is a lot smaller than the benchmark system
(120000 atoms) scaling will be worse as well. Check out the benchmark...
Dr. David van der Spoel, Biomedical center, Dept. of Biochemistry
Husargatan 3, Box 576, 75123 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