[Mauiusers] fairshare and backfill
csamuel at vpac.org
Sat Feb 23 20:57:08 MST 2008
----- "Jim Kusznir" <jkusznir at gmail.com> wrote:
> On the note of this thread, it is also my understanding that
> maui/torque on linux is not able to preempt jobs. That is to say,
> once a job has been assigned resources and starts running, there is
> no way to "suspend/pause" the job.
I was under the impression that some people had this
working a while ago - basically the job gets sent
SIGSTOP by pbs_mom (more complicated than that, but
you get the idea).
There is also preliminary support for the BCLR
checkpoint/restart kernel module in trunk for Torque
that will checkpoint the job to disk and terminate it,
thus freeing up any resources it was using.
Christopher Samuel - (03) 9925 4751 - Systems Manager
The Victorian Partnership for Advanced Computing
P.O. Box 201, Carlton South, VIC 3053, Australia
VPAC is a not-for-profit Registered Research Agency
More information about the mauiusers