[torqueusers] Resources_List.ncpus & resources_assigned.ncpus
csamuel at vpac.org
Mon Sep 5 17:57:00 MDT 2005
On Tue, 30 Aug 2005 04:13 pm, Garrick Staples wrote:
> On Tue, Aug 30, 2005 at 02:27:44PM +1000, Chris Samuel alleged:
> > It would *wonderful* if it could be turned into a way to request a job
> > that wants 'n' cpus in *any* configuration..
> Chris! Wake up! You're dreaming again.
Actually, now that I've realised that setting resources_available.nodect to
the number of CPUs in the cluster got fixed somewhere between 1.2.0p1 and
1.2.0p5 we can use "nodes=" to do this again.
The problem before was that without that you couldn't just ask for X>N cpus
where N was the number of nodes in the cluster via the "nodes=" spec.
I've closed bug #25 that I'd put in about this; many thanks to the fixer!
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
Size: 189 bytes
Desc: not available
Url : http://www.supercluster.org/pipermail/torqueusers/attachments/20050906/1eef554c/attachment.bin
More information about the torqueusers