[gmx-users] How to set up the cut-off radius

[Yinhe Zhang]

Hi, Yanmei

On Fri, May 8, 2009 at 2:20 AM, Yanmei Song <ysong30 at asu.edu> wrote:

> NOTE 2 [file aminoacids.dat, line 1]:
>   The optimal PME mesh load for parallel simulations is below 0.5
>   and for highly parallel simulations between 0.25 and 0.33,
>   for higher performance, increase the cut-off and the PME grid spacing
>

about note 2, the info given says if you run mdrun in parallel machine you
may better
make PME mesh load to below 0.5 to get optimal performace. If you run it
just in a one-node
machine, you may just ignore the note. But if in the first case of running
in parallel, you can increase fft grid spacing a little (just try several
values little larger)to lower PME mesh load to better distribute the
computation among nodes.

Regards,
Yinhe Zhang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090508/37bc70ad/attachment.html>