Re: [AMBER] AMBER Digest, Vol 1973, Issue 1

From: David Case <david.case.rutgers.edu>
Date: Thu, 29 Jun 2017 08:35:50 -0600

On Thu, Jun 29, 2017, Garima Singh wrote:
>
> I am using fewer nodes this time using
> below pbs script. Job run and and still continue but it only give 7 ns
> /day which is very slow.How to get fast result from HPC.

> mpirun -np 160 sander.MPI ....

160 threads is still *way* to many for sander! Sander will likely max out
at something like 8 to 16 threads. Run a bunch of *short* simulations with
various thread counts to gain some experience with performance.

> Is it slow because of Sander module.Can i use
> PMEMD to get fast calculation using HPC.

pmemd will generally be faster than sander, and will scale better to large
numnber of threads. But don't be afraid to experiment: find out what
pmemd will do on your system, again testing performance vs. number of threads.

...good luck....dac


_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jun 29 2017 - 08:00:02 PDT
Custom Search