[torqueusers] vmem

Gareth.Williams at csiro.au Gareth.Williams at csiro.au
Wed Nov 13 15:32:19 MST 2013


Or just keep doubling the vmem request and resubmitting the job until it works (no admin action required). The tune the request back down.

Gareth

From: David Beer [mailto:dbeer at adaptivecomputing.com]
Sent: Thursday, 14 November 2013 3:41 AM
To: Torque Users Mailing List
Subject: Re: [torqueusers] vmem

Delphine,

I'm not certain how to get a good value for vmem other than through empirical use. One possible way to do this would be to set

$ignvmem true

in the mom's config file and reboot the mom. This parameter makes it so that the mom will not kill the job for going over the limit. Then you could see about how much the job uses and make your requests based on that.

David

On Wed, Nov 13, 2013 at 6:09 AM, Delphine Ramalingom <delphine.ramalingom at univ-reunion.fr<mailto:delphine.ramalingom at univ-reunion.fr>> wrote:
Hi,

I have a problem with vmem. How can I know the value I should fix to
vmem in my script ?
I put
#PBS -l nodes=1:ppn=1,mem=28gb,vmem=28gb

and I have the error message :
=>> PBS: job killed: vmem 37569708032 exceeded limit 30064771072

We have Torque + maui installed and configured to cancel job that exceed
limits.

Regards
Delphine
_______________________________________________
torqueusers mailing list
torqueusers at supercluster.org<mailto:torqueusers at supercluster.org>
http://www.supercluster.org/mailman/listinfo/torqueusers



--
David Beer | Senior Software Engineer
Adaptive Computing
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20131113/65546172/attachment.html 


More information about the torqueusers mailing list