[slurm-users] Performance tracking of array tasks
toomuchit at gmail.com
Mon May 16 13:56:11 UTC 2022
I think you may be conflating the array task with slurmstepd tasks.
Every array task does have it's own job id, usually <JOBID>_<TASKID>
It will also have a 'normal' job id that goes with that.
You can get detailed information from sacct if you have configured it.
In fact, it also breaks things down into slurm steps, so you would be
able to get pretty good detailed info.
On 5/16/2022 6:44 AM, William Dear wrote:
> Could anyone please recommend methods of tracking the performance of
> individual tasks in a task array job? I have installed XDMoD but it
> is focused solely on the Job level with no information about tasks.
> My users almost exclusively use task arrays to run embarrassingly
> parallel jobs. After the job is complete I would like to see run time
> and peak RAM usage per task so that we can correctly size the
> reservations for future jobs. It would also be very helpful to break
> this down by node so that I can identify poorly performing nodes.
> William Dear
> *Confidentiality Notice* - The information contained in this e-mail
> and any attachments to it may be legally privileged and include
> confidential information. If you are not the intended recipient, be
> aware that any disclosure, distribution or copying of this e-mail or
> its attachments is prohibited. If you have received this e-mail in
> error, please notify the sender immediately of that fact by return
> e-mail and permanently delete the e-mail and any attachments to it.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the slurm-users