problem with multi-gpu on CUDA & OpenCL

The functionality of OpenMM will (eventually) include everything that one would need to run modern molecular simulation.
User avatar
Peter Eastman
Posts: 2583
Joined: Thu Aug 09, 2007 1:25 pm

Re: problem with multi-gpu on CUDA & OpenCL

Post by Peter Eastman » Mon Jun 06, 2016 3:17 pm

I found the problem. It's wasn't actually related to multiple GPUs. It was caused by the fact that the Drude particles were at exactly the same positions as their parent particles. That led to a nan in the force computation. The fix is at https://github.com/pandegroup/openmm/pull/1509. Could you give it a try?

Peter

User avatar
Mary Van Vleet
Posts: 18
Joined: Wed Dec 09, 2015 11:56 am

Re: problem with multi-gpu on CUDA & OpenCL

Post by Mary Van Vleet » Tue Jun 07, 2016 8:01 am

Peter,

Yes, this fixes my problem! As always, thank you so much for your timely help.

Cheers,
Mary

User avatar
Peter Eastman
Posts: 2583
Joined: Thu Aug 09, 2007 1:25 pm

Re: problem with multi-gpu on CUDA & OpenCL

Post by Peter Eastman » Tue Jun 07, 2016 8:18 am

Ok, great. The fix is merged.

Peter

POST REPLY