[CPMD-list] Running CPMD on multi-core Opeton cluster

[Kozin, I (Igor)]

There are many ways you can go about this especially if you are submitting via a queue system
For example you can edit the machine file and repeat every node four times.
Booting up 32 and requesting 128 should work too because mpd works in a round robin fashion.
 
 

-----Original Message-----
From: cpmd-list-bounces at cpmd.org [mailto:cpmd-list-bounces at cpmd.org]On Behalf Of Yiming Zhang
Sent: 31 October 2006 18:44
To: cpmd-list at cpmd.org
Subject: [CPMD-list] Running CPMD on multi-core Opeton cluster


Dear CPMD community,

I am currently running CPMD on a Opteron AMD64 cluster, every node has 4 CPU (2 dual core Opterons).
I am able to parallel CPMD using MPI MPD, an example script is pasted below. When I ask for 32 for $NSLOTS. it actually runs on 32 nodes with only 1 of 4 cpus per node running. I am wondering if there is a way to fully utilize the cluster, running on all cpus on 32 nodes, then totally I can have 128 cpus. 

If I simply ask for larger $NSLOTS, the MPD quit with error message like "more processors asked than available", I can't remember exactly.

Any suggestions are welcome,
Yiming Zhang


mpdboot -n $NSLOTS -f $MACHINEFILE 
mpdtrace -l
mpiexec -n $NSLOTS /borg/yiming/borg_cpmd.x $base/inp > $base/out
mpdallexit




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://cpmd.org/pipermail/cpmd-list/attachments/20061031/2801f40f/attachment.html