[Moabusers] soft limits on MAXIJOB
Douglas Wightman
wightman at clusterresources.com
Thu Apr 5 16:23:51 MDT 2007
Hard and soft limits only apply to active job policies.
Have you tried configuring limits on the individual queues?
CLASSCFG[batch] MAXIJOB[USER]=2
- Douglas
On Wed, 2007-04-04 at 10:58 -0600, kamil Marcinkowski wrote:
> Hello All,
>
>
> We are running a "large SMP" cluster with short max wall-time
> and a few queues (classes) mapped. Each queue mapped
> to exclusive hosts. We want to implement anti
> queue-stuffing measures.
>
>
> If we set MAXIJOB 2, no more than 2 jobs per user will acquire
> queue-time however only first 2 jobs of an user will be considered
> for running. This is a problem since the first two jobs of a user can
> be in
> one queue with lot of other jobs and other jobs of the user can be
> in a queue that is otherwise empty, leaving an empty machine.
>
>
> Moab admin manual is unclear wether soft limits on
> MAXIJOB are supported.
> We tried setting soft MAXIJOB but it did does not see to work,
> queue stuffed jobs are still gaining queue-time.
>
>
> > USERCFG[DEFAULT] MAXIJOB=2,8
>
>
> We are running:
> Moab Workload Manager version 4.5.0p3
> Torque version: 2.1.6
>
>
> Is this supposed to work?
> If not is there an alternative way of achieving the same results?
>
>
> Thanks,
>
>
> Kamil
>
> Kamil Marcinkowski Westgrid System Administrator
> kamil at ualberta.ca University of Alberta site
>
> Tel.780 492-0354 Research Computing Support
> Fax.780 492-1729 Academic ICT
> Edmonton, Alberta, CANADA University of Alberta
>
>
>
>
> "This communication is intended for the use of the recipient to which
> it is
> addressed, and may contain confidential, personal, and/or privileged
> information. Please contact us immediately if you are not the
> intended
> recipient of this communication. If you are not the intended
> recipient of
> this communication, do not copy, distribute, or take action on it. Any
> communication received in error, or subsequent reply, should be
> deleted or
> destroyed."
>
>
>
>
>
>
>
> _______________________________________________
> moabusers mailing list
> moabusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/moabusers
More information about the moabusers
mailing list