Dear amber users,
I am quite new to amber (version16 used) and unix in general, it would be great to get an answer to the following questions.
I am running calcs on a cluster. Up to now I used for sander MPI parallelisable jobs 32 cpu’s. However no real cpu’s since they are up
to 4 times overbooked. Today one calculation took 10 times longer then normal. However support asked me if I wanted to have more
cpu’s – so my question is up to which point does it make sense to enlarge the number of cpu’s. My system contains peptidomimetic with 12 residues
in DMSO or TIP4PEW.
Furthermore I run on 1 GPU all pmemd calcs. However I was asked whether I want to have another GPU to finish the project earlier.
I read online:
Support is provided for single GPU and multiple GPU runs. Employing multiple GPUs in a single simulation requires MPI and the pmemd.cuda.MPI executable.
If you have multiple simulations to run then the recommended method is to use one GPU per job.
So I want to run 2 jobs in parallel, do I need to specify for each job which GPU to be addressed, or is pmemd by default taking non useds GPU’s first?
In the mailing list I found: http://archive.ambermd.org/201706/0234.html
However as a newbie I don’t understand everything here.
Thanks for explaining me,
Best regards,
David
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Aug 07 2018 - 10:00:02 PDT
