The specific issue that I want to address is: "My job runs slower when there are other jobs on the same node and some of the nodes are unused, why doesn't my job run on the unused nodes?"
I don't want to foster the habit of scheduling jobs with --nodelist P.S. We currently have fairly low cluster utilization so some nodes really do sit idle most of the time. Wasteful, yes, but there is so much politics wrapped up in that that I don't want to go there right now. On Thu, May 8, 2014 at 11:00 AM, Ryan Cox <[email protected]> wrote: > Rather than maximize fragmentation, you probably want to do it on a > per-job basis. If you want one core per node: sbatch: -N $numnodes -n > $numnodes. Anything else would require the -m flag. I haven't played with > it recently but I think you would want -m cyclic. > > Ryan > > > On 05/08/2014 11:49 AM, Atom Powers wrote: > > > It appears that my Slurm cluster is scheduling jobs to load up nodes as > much as possible before putting jobs on other nodes. I understand the > reasons for doing this, however I foresee my users wanting to spread jobs > out among as many nodes as possible for various reasons, some of which are > even valid. > > How would I configure the scheduler to distribute jobs in something like > a round-robin fashion to many nodes instead of loading jobs onto just a few > nodes? > > I currently have: > 'SchedulerType' => 'sched/builtin', > 'SelectTypeParameters' => 'CR_Core_Memory', > 'SelectType' => 'select/cons_res', > > -- > Perfection is just a word I use occasionally with mustard. > --Atom Powers-- > > > -- > Ryan Cox > Operations Director > Fulton Supercomputing Lab > Brigham Young University > > -- Perfection is just a word I use occasionally with mustard. --Atom Powers--
