Hi Herbert,

I believe the limit is per node (not per partition) whereas you want it per job. In other words, your users will be able to run jobs on other nodes.

There is no MaxCPUsPerJob option in the partition definition, but I believe you can make that restriction in other ways (at worst with a job_submit.lua but I think there would be a simpler way).

Sorry not an answer, but hopefully a little nudge toward the solution

Davide

On Wed, Feb 26, 2025 at 11:37 AM Herbert Fruchtl via slurm-users <slurm-users@lists.schedmd.com> wrote:

We have a cluster with multi-core nodes (168) that can be shared by
multiple jobs at the same time. How do I configure a partition such that
it only accepts jobs requesting up to (say) 8 cores, but will run
multiple jobs at the same time? The following is apparently not working:

PartitionName=debug Nodes=node01 MaxTime=02:00:00 DefMemPerCPU=1000
MaxCPUsPerNode=8 Default=NO

It allows one job using 8 cores, but a second one will not start because
the limit is apparently for the partition as a whole.

Thanks in advance,

Herbert
--
Herbert Fruchtl (he/him)
Senior Scientific Computing Officer / HPC Administrator
School of Chemistry, IT Services
University of St Andrews
--
The University of St Andrews is a charity registered in Scotland:
No SC013532

--
slurm-users mailing list -- slurm-users@lists.schedmd.com
To unsubscribe send an email to slurm-users-leave@lists.schedmd.com