[torqueusers] general questions about installation and submitting job

[Chris Samuel]

On Monday 11 December 2006 23:04, Ronny T. Lampert wrote:

> You specified 2 _NODES_, not _CPUs_ in your job requirement.
> As you only have 1 node torque is right to reject that job.

This is the old chestnut that there is no simple way to request an allocation 
of N cpus anywhere in the cluster through Torque.  I'd suggested a possible 
option of:

 -l cpus=N 

but that would require changes in pbs_sched, Maui and Moab to support it.

The kludge around this is to set, through qmgr:

 set server resources_available.nodect = N

where N is the total number of CPUs in the cluster.

-- 
 Christopher Samuel - (03)9925 4751 - VPAC Deputy Systems Manager
 Victorian Partnership for Advanced Computing http://www.vpac.org/
 Bldg 91, 110 Victoria Street, Carlton South, VIC 3053, Australia

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://www.supercluster.org/pipermail/torqueusers/attachments/20061222/7eb3c836/attachment.bin