Re: [AMBER] Running Amber 11 simulations using pmemd.cuda.MPI

From: Ross Walker <>
Date: Wed, 6 Jul 2011 08:54:42 -0700

Hi David,

> Thank you for your advice. I've spend the day working on mvapich2 and
> Amber. Rebuilding pmemd.cuda.MPI using mvapich2 is exactly what's
> needed. Using 4 GPUs (that is two compute nodes) on the
> PME/Cellulose_production_NPT benchmark example I can get the simulation
> done in 15 minutes. This is excellent scaling since the same simulation
> takes 30 mins using 2 GPUs.

This is excellent news. I'm glad you got it working. To give you and update
I am hoping to have a patch out very soon (just working out the last of the
bugs) that will double the performance over what you currently are getting.
So then hopefully you'll be down around 8 minutes or so.
> I'll need to open this testing up to some of the Amber users here
> before we can it a success. Pity that Openmpi doesn't do the job -- I'm
> not that keen to have to offer another flavor of MPI2 on the cluster.
> Taking a look at the latest version of OpenMPI, off hand, it appears
> that they are no way close to supporting GPUs properly.

To be honest I would just ditch openMPI completely and make mvapich2 the
only MPI installation on the cluster. It will benefit even CPU codes
especially if you go to larger numbers of nodes. The only downside is people
will need to recompile but it should be fairly simple to do.

All the best

|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Adjunct Assistant Professor |
| Dept. of Chemistry and Biochemistry |
| University of California San Diego |
| NVIDIA Fellow |
| | |
| Tel: +1 858 822 0854 | EMail:- |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

AMBER mailing list
Received on Wed Jul 06 2011 - 09:00:04 PDT
Custom Search