Re: [AMBER] pmemd.cuda issues on Tesla S2050

From: Scott Le Grand <>
Date: Thu, 10 Feb 2011 06:51:21 -0800

In that new code, there is a possible kludge to work around this. Go into gputypes.h and remove the #ifdef use_DPDP from around NODPTEXTURE.

If that either fixes or makes the problem less common, it is indeed the same issue as on 480s and 580s...

-----Original Message-----
From: Ross Walker []
Sent: Wednesday, February 09, 2011 22:37
To: 'AMBER Mailing List'
Subject: Re: [AMBER] pmemd.cuda issues on Tesla S2050

> The Amber11 freezing occurs on Teslas with any number of GPUs - it's
> happened on 6- and 8-card systems as well. Also, these lockups don't
> appear to be associated with the other problem. When pmemd.cuda locks
> up, the system still recognizes all installed GPUs.
> Can you think of a good way to deal with it? Before we dive into
> complicated wrapper scripts with timeouts..

We have some new (and much more optimized code) that I am currently testing.
Email me off list and you can try that out and see if you still see the
lockup problem.

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
This email message is for the sole use of the intended recipient(s) and may contain
confidential information. Any unauthorized review, use, disclosure or distribution
is prohibited. If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.

AMBER mailing list
Received on Thu Feb 10 2011 - 07:00:04 PST
Custom Search