On 10/5/22 22:43, Carl Ponder wrote:
In our scaling tests, it's normal to expect the job run-times to reduce as
we increase the node-counts.
Is there a way in SLURM to limit the NODES*TIME product for a partition,
or do we just have to define a different partition (with a different
duration-limit) for each job size?
AFAIK this is not possible, see the slurm.conf section
https://slurm.schedmd.com/slurm.conf.html#SECTION_PARTITION-CONFIGURATION
for available partition parameters.
Limits of nodes*time can, however, be defined on user accounts, see
GrpTRESRunMins in https://slurm.schedmd.com/resource_limits.html
Short jobs can run in a partition with a higher MaxTime than required. I
don't see why you would want separate partitions with different MaxTimes?
/Ole
