BwUniCluster3.0/Running Jobs - Revision history

S Braun: /* Interactive Computing with Jupyter */

2026-02-03T14:36:30Z

Interactive Computing with Jupyter

← Older revision		Revision as of 16:36, 3 February 2026
Line 414:		Line 414:

	If you are using Intel MPI you must start <my_mpi_program> by the command mpiexec.hydra (instead of mpirun).		If you are using Intel MPI you must start <my_mpi_program> by the command mpiexec.hydra (instead of mpirun).

	== Interactive Computing with Jupyter ==

	== Monitor and manage jobs ==		== Monitor and manage jobs ==

2026-02-03T14:35:52Z

Monitor and manage jobs

← Older revision		Revision as of 16:35, 3 February 2026
Line 438:		Line 438:

	</pre>		</pre>


			If resources are not immediately available add <code>--start</code> to show its expected start time:

			<syntaxhighlight lang="sh">squeue --start</syntaxhighlight>



	=== Detailed job information : scontrol show job ===		=== Detailed job information : scontrol show job ===

2026-01-07T13:52:09Z

Batch Jobs: sbatch

← Older revision		Revision as of 15:52, 7 January 2026
Line 266:		Line 266:
	\| #SBATCH --cpus-per-task=''count''		\| #SBATCH --cpus-per-task=''count''
	\| Number of CPUs required per (MPI-)task.		\| Number of CPUs required per (MPI-)task.
			\|-
			\|- style="vertical-align:top;"
			\| --gres=gpu:''count''
			\| #SBATCH --gres=gpu:''count''
			\| Number of GPUs required per node.
	\|-		\|-
	\|- style="vertical-align:top;"		\|- style="vertical-align:top;"

2025-10-24T08:07:44Z

← Older revision		Revision as of 10:07, 24 October 2025
Line 483:		Line 483:
	</pre>		</pre>
	<br>		<br>
	Each request to the Slurm workload manager generates a load. <p style="color:red;> Therefore, do not use <code>squeue</code> with a simple <code>watch</code>.</p> The smallest allowed time interval is <b>30 seconds</b>.<br>		Each request to the Slurm workload manager generates a load. <p style="color:red;><b>Therefore, do not use <code>squeue</code> with a simple <code>watch</code>.</b></p> The smallest allowed time interval is <b>30 seconds</b>.<br>
	Any violation of this rule will result in the task being terminated without notice.		Any violation of this rule will result in the task being terminated without notice.

2025-10-24T08:06:40Z

Detailed job information : scontrol show job

← Older revision		Revision as of 10:06, 24 October 2025
Line 483:		Line 483:
	</pre>		</pre>
	<br>		<br>
	Each request to the Slurm workload manager generates a load. Therefore, do not use <code>squeue</code> with a simple <code>watch</code>. The smallest allowed time interval is <b>30 seconds</b>.<br>		Each request to the Slurm workload manager generates a load. <p style="color:red;> Therefore, do not use <code>squeue</code> with a simple <code>watch</code>.</p> The smallest allowed time interval is <b>30 seconds</b>.<br>
	Any violation of this rule will result in the task being terminated without notice.		Any violation of this rule will result in the task being terminated without notice.

2025-10-24T08:05:14Z

Detailed job information : scontrol show job

← Older revision		Revision as of 10:05, 24 October 2025
Line 443:		Line 443:
	<br>		<br>

	* Here is an example from bwUniCluster 3.0.		Here is an example from bwUniCluster 3.0:
	<pre>		<pre>
	$ squeue		$ squeue
Line 483:		Line 483:
	</pre>		</pre>
	<br>		<br>
			Each request to the Slurm workload manager generates a load. Therefore, do not use <code>squeue</code> with a simple <code>watch</code>. The smallest allowed time interval is <b>30 seconds</b>.<br>
			Any violation of this rule will result in the task being terminated without notice.

	=== Canceling own jobs : scancel ===		=== Canceling own jobs : scancel ===
	The scancel command is used to cancel jobs. The command scancel is explained in detail on the webpage https://slurm.schedmd.com/scancel.html or via manpage (man scancel). The command is:		The scancel command is used to cancel jobs. The command scancel is explained in detail on the webpage https://slurm.schedmd.com/scancel.html or via manpage (man scancel). The command is:

2025-09-23T15:08:16Z

Dos and Don'ts

← Older revision		Revision as of 17:08, 23 September 2025
Line 498:		Line 498:

	== Dos and Don'ts ==		== Dos and Don'ts ==

			{\|style="background:#deffee; width:100%;"
			\|style="padding:5px; background:#cef2e0; text-align:left"\|
			[[Image:Attention.svg\|center\|25px]]
			\|style="padding:5px; background:#cef2e0; text-align:left"\| Do not run squeue and other slurm commands in loops or "watch" as not to saturate up the slurm daemon with rpc requests
			\|}

2025-08-19T09:21:20Z

Batch Jobs: sbatch

← Older revision		Revision as of 11:21, 19 August 2025
Line 276:		Line 276:
	\| #SBATCH --mem-per-cpu=''value_in_MB''		\| #SBATCH --mem-per-cpu=''value_in_MB''
	\| Minimum Memory required per allocated CPU. (You should omit the setting of this option.)		\| Minimum Memory required per allocated CPU. (You should omit the setting of this option.)
			\|-
			\|- style="vertical-align:top;"
			\| --exclusive
			\| #SBATCH --exclusive
			\| The job allocates all CPUs and GPUs on the nodes. It will not share the node with other running jobs
	\|-		\|-
	\|- style="vertical-align:top;"		\|- style="vertical-align:top;"

2025-08-13T12:35:24Z

Regular Queues

← Older revision		Revision as of 14:35, 13 August 2025
Line 110:		Line 110:
	\| mem-per-gpu=127500mb<br/>cpus-per-gpu=16		\| mem-per-gpu=127500mb<br/>cpus-per-gpu=16
	\| gres=gpu:1		\| gres=gpu:1
	\| time=48:00:00, nodes=9(A100)\|nodes=5(H100) , mem=510000mb, ntasks-per-node=64, (threads-per-core=2)		\| time=48:00:00, nodes=9(A100)/nodes=5(H100) , mem=510000mb, ntasks-per-node=64, (threads-per-core=2)
	\|}		\|}
	Table 1: Regular Queues		Table 1: Regular Queues

2025-08-13T12:35:08Z

Regular Queues

← Older revision		Revision as of 14:35, 13 August 2025
Line 110:		Line 110:
	\| mem-per-gpu=127500mb<br/>cpus-per-gpu=16		\| mem-per-gpu=127500mb<br/>cpus-per-gpu=16
	\| gres=gpu:1		\| gres=gpu:1
	\| time=48:00:00, nodes=9(A100) nodes=5(H100) , mem=510000mb, ntasks-per-node=64, (threads-per-core=2)		\| time=48:00:00, nodes=9(A100)\|nodes=5(H100) , mem=510000mb, ntasks-per-node=64, (threads-per-core=2)
	\|}		\|}
	Table 1: Regular Queues		Table 1: Regular Queues