BwForCluster JUSTUS 2 Slurm HOWTO - Revision history

H Schumacher: added interactive job cancelling

2026-07-30T17:06:49Z

added interactive job cancelling

← Older revision		Revision as of 19:06, 30 July 2026
Line 690:		Line 690:

	<pre>		<pre>
	$ scancel <jobid> # cancel specific job		$ scancel <jobid> # cancel specific job
	$ scancel <jobid>_<index> # cancel indexed job in a job array		$ scancel <jobid>_<index> # cancel indexed job in a job array
	$ scancel -u <username> # cancel all jobs of specific user		$ scancel -u <username> # cancel all jobs of specific user
	$ scancel -t ~~PENDING~~ # ~~cancel~~ ~~pending~~ ~~jobs~~		$ scancel -u <username> -i # cancel all jobs of specific user interactively
	$ scancel -t ~~RUNNING~~ # cancel ~~running~~ jobs		$ scancel -t PENDING # cancel pending jobs
			$ scancel -t RUNNING # cancel running jobs
	</pre>		</pre>

2025-09-30T17:35:40Z

← Older revision		Revision as of 19:35, 30 September 2025
Line 409:		Line 409:
	# module load foo/bar		# module load foo/bar

	export OMP_NUM_THREADS=${~~SLURM_JOB_CPUS_PER_NODE~~}		export OMP_NUM_THREADS=${SLURM_CPUS_PER_TASK}
	export MKL_NUM_THREADS=${~~SLURM_JOB_CPUS_PER_NODE~~}		export MKL_NUM_THREADS=${SLURM_CPUS_PER_TASK}

	# Run multithreaded program		# Run multithreaded program

2025-03-13T08:02:31Z

How to convert Moab batch job scripts to Slurm?

← Older revision		Revision as of 10:02, 13 March 2025
Line 147:		Line 147:
	\| Process count per node \|\| -l ppn=<count> \|\| --ntasks-per-node=<count>		\| Process count per node \|\| -l ppn=<count> \|\| --ntasks-per-node=<count>
	\|-		\|-
	\| Core count per process \|\| \|\| --cpus-per-task=<count>		\| Core count per process \|\| \|\| --cpus-per-task=<count> (-c <count>)
	\|-		\|-
	\| Memory limit per node \|\| -l mem=<limit> \|\| --mem=<limit>		\| Memory limit per node \|\| -l mem=<limit> \|\| --mem=<limit>

2024-10-23T15:21:00Z

← Older revision		Revision as of 17:21, 23 October 2024
Line 1,117:		Line 1,117:

	<pre>		<pre>
	$ sacctmgr modify user where user=<username> set DefaultAccount=<default_account>		$ sacctmgr modify user where user=<username> set DefaultAccount=<default_account>
	</pre>		</pre>

2024-10-23T15:19:01Z

← Older revision		Revision as of 17:19, 23 October 2024
Line 1,114:		Line 1,114:
	</pre>		</pre>

	== How to change default account of a user?		== How to change default account of a user? ==

	<pre>		<pre>

2024-10-23T15:17:28Z

← Older revision		Revision as of 17:17, 23 October 2024
Line 1,113:		Line 1,113:
	$ sacctmgr remove user <username> where account=<accountname> # Remove user from this account		$ sacctmgr remove user <username> where account=<accountname> # Remove user from this account
	</pre>		</pre>

			== How to change default account of a user?

			<pre>
			$ sacctmgr modify user where user=<username> set DefaultAccount=<default_account>
			</pre>

			'''Note:''' The user must already be associated with the account you want to set as default.

	== How to show account information? ==		== How to show account information? ==

2022-12-02T15:34:44Z

How to find working sample scripts for my program?

@@ Line 82: / Line 82: @@
 $ module help math/matlab   # display module help for Matlab
 </pre>
 == How to submit an interactive job? ==

2022-09-20T23:34:07Z

How to submit a multithreaded batch job?

← Older revision		Revision as of 01:34, 21 September 2022
Line 408:		Line 408:

	* In our configuration each physical core is considered a "CPU".		* In our configuration each physical core is considered a "CPU".

			* On JUSTUS 2 it is recommended to specify a number of cores per task ('--cpus-per-task') that is either an integer divisor of 24 or (at most) 48.

	* Required memory can also by specified per allocated CPU with '--mem-per-cpu' option.		* Required memory can also by specified per allocated CPU with '--mem-per-cpu' option.

	* The '--mem' and '--mem-per-cpu' options are mutually exclusive.		* The '--mem' and '--mem-per-cpu' options are mutually exclusive.

	* In terms of core allocation '--ntasks-per-node=1' or '--ntasks=1' together with '--cpus-per-task=8' is almost equivalent to '--ntasks-per-node=8' or '--ntasks=8' and omitting '--cpus-per-task=8'. However, there are subtle differences when multiple tasks are spawned within one job by means of srun command.
	** See: https://stackoverflow.com/questions/39186698/what-does-the-ntasks-or-n-tasks-does-in-slurm

	== How to submit an array job? ==		== How to submit an array job? ==

2022-08-15T15:37:24Z

How to retrieve historical resource usage records for a specific user or account?

← Older revision		Revision as of 17:37, 15 August 2022
Line 1,138:		Line 1,138:
	</pre>		</pre>

	== How to retrieve historical resource usage ~~records~~ for a specific user or account? ==		== How to retrieve historical resource usage for a specific user or account? ==

	Use [https://slurm.schedmd.com/sreport.html sreport] command.		Use [https://slurm.schedmd.com/sreport.html sreport] command.

2022-08-15T15:33:13Z

How to retrieve historical resource usage records for a specific user?

← Older revision		Revision as of 17:33, 15 August 2022
Line 1,138:		Line 1,138:
	</pre>		</pre>

	== How to retrieve historical resource usage records for a specific user? ==		== How to retrieve historical resource usage records for a specific user or account? ==

	Use [https://slurm.schedmd.com/sreport.html sreport] command.		Use [https://slurm.schedmd.com/sreport.html sreport] command.