AW: Simultaneous calculation on CPU-only nodes and CPU/GPU node (with or without rCUDA)

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Tue May 29 2012 - 01:05:16 CDT

Try fullelectfrequency 4, harms energy conservation but speedups the
simulation!

Norman Geist.

> -----Ursprüngliche Nachricht-----
> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> Auftrag von Benjamin Merget
> Gesendet: Sonntag, 27. Mai 2012 16:11
> An: Axel Kohlmeyer
> Cc: namd-l_at_ks.uiuc.edu
> Betreff: Re: namd-l: Simultaneous calculation on CPU-only nodes and
> CPU/GPU node (with or without rCUDA)
>
> The problem is, that I can only reach up to about 25% gpu utilization
> of
> each of the 4 Tesla cards. I thought that maybe I could increase the
> GPU
> utilization by creating more processes to bind to the Tesla cards. But
> to do so, I need more CPUs, i.e. the CPUs of my CPU-only nodes...
>
> Is there another way to increase my GPU utilization with the 8 CPU
> cores
> of my GPU node?
>
> Benny
>
> > no, and it is not worth it. just run one calculation on the GPU node
> > and a second on the rest and enjoy efficient utilization of your
> hardware.
> > anything else is just wasting your time.
> >
> > axel.
> >

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:21:34 CST