-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 24/07/13 03:00, Tim Wickberg wrote:

> I'm assuming the jobs are running across multiple nodes, using MPI
> for communication?

That's correct, 70 nodes each running 16 ranks per node.

> I'm guessing that srun is resulting in communication going across a
> GigE fabric rather than IB, where mpirun directly is using the IB.
> A ~20% performance penalty would make sense in that context.

Sadly I don't believe that's the case.  I've checked and the NAMD
processes have the IB verbs devices open in the same way they do with
mpirun, plus I believe that NAMD2 is quite closely coupled as in the
past we've run it over gigE as a test and found it performs really
poorly.  That was quite a few years ago now though since last time we
tried it.

All the best,
Chris
- -- 
 Christopher Samuel        Senior Systems Administrator
 VLSCI - Victorian Life Sciences Computation Initiative
 Email: sam...@unimelb.edu.au Phone: +61 (0)3 903 55545
 http://www.vlsci.org.au/      http://twitter.com/vlsci

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlHvEikACgkQO2KABBYQAh/hHQCeLmZBX0YOYgVso+/P+lSkt8eZ
8noAn1NBHOiJF0Eww6+TX32na0CDVHVk
=mkb0
-----END PGP SIGNATURE-----

Reply via email to