Re: [AMBER] cuda job not working

From: Daniel Roe <daniel.r.roe.gmail.com>
Date: Sat, 4 Jan 2014 11:14:47 -0700

Hi,


On Sat, Jan 4, 2014 at 8:59 AM, Jio M <jiomm.yahoo.com> wrote:

> 1) I get this: so looks like its not updated?
>

No; this code is actually way out of date and you need to update. Simply
re-run configure, apply all available patches, and re-compile the code.


> 3) Another thing I noted: job needs to be exclusive in a node. I use bsub
> submission scripts and with that -x flag allows to use node exclusively for
> one job but when I don't use this and submit on free GPU in some node with
> some job already running on it, my AMBER job kills with same error.
>

I'm not 100% sure, but it sounds like whatever queuing software you are
using isn't properly allocating GPUs by e.g. setting CUDA_VISIBLE_DEVICES.
An Amber CUDA job should have exclusive access to the GPUs it is using. It
is not recommended for jobs to share GPUs; this can lead to unexpected and
bad behavior. If you haven't already, see
http://ambermd.org/gpus/#Runningfor more information on running CUDA
jobs.

Hope this helps,

-Dan

-- 
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jan 04 2014 - 10:30:02 PST
Custom Search