David L. Bostick wrote:
Dear all,
I have done some testing on the two different Xeon clusters. For clarity
<snippage>
Clearly machine (2) does not scale at all when using PME. I suspected it
was due to fftw3, and compiled using intel mkl, but it doesn't appear to behave
differently. Right now I am at a loss for an explanation, and would
appreciate some help.
OK, it seems you've got a real problem - you've demonstrated correct
non-PME parallel scaling behaviour, so the problem seems not to be with
your command lines, parallel gromacs install, MPI install, MPI config,
or networking hardware.
How about comparing runs that don't use PME, and thus don't use fftw? I
can't really see the kernel version having any impact... you could compare
the output from ./configure to see if anything else significant is
different. The most likely problem is not using the same number of
processors, but you say you've done that! :-) You could also check whether
there is a horrible MPI latency problem with machine (2)
It's still worth checking the gromacs ./configure output for meaningful
differences.
You could try looking for parallel FFTW benchmarks to see if the problem
can be isolated there. Perhaps their source distribution includes a test
suite, or Google might be your friend here.
Mark
_______________________________________________
gmx-users mailing list gmx-users@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 [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php
