How to use array jobs in SLURM - Revision history

Dmitri: /* Warning */

2024-11-28T22:33:41Z

Warning

← Older revision		Revision as of 22:33, 28 November 2024
Line 6:		Line 6:
	* Please, '''first time''' submit '''only one job of the set''' as a test case and make sure that it runs and completes successfully.		* Please, '''first time''' submit '''only one job of the set''' as a test case and make sure that it runs and completes successfully.
	* Then submit an '''array of 2-3 jobs''', to make sure that you fully understand how array will be working.		* Then submit an '''array of 2-3 jobs''', to make sure that you fully understand how array will be working.
	* Only then, commit to submitting the '''full data set''' ~~processing as the complete array~~.		* Only then, commit to submitting the array job for the '''full data set'''.

Dmitri: /* Warning */

2024-11-28T22:32:24Z

Warning

← Older revision		Revision as of 22:32, 28 November 2024
Line 2:		Line 2:
	Job arrays is an '''advanced technique'''. Some '''caution should be used''' when submitting array of jobs.		Job arrays is an '''advanced technique'''. Some '''caution should be used''' when submitting array of jobs.

	Please never submit a full set of jobs when starting a new set of calculation.
			Please '''never submit a full set''' of jobs when starting a '''new set of calculation'''.
	* Please, '''first time''' submit '''only one job of the set''' as a test case and make sure that it runs and completes successfully.		* Please, '''first time''' submit '''only one job of the set''' as a test case and make sure that it runs and completes successfully.
	* Then submit an '''array of 2-3 jobs''', to make sure that you fully understand how array will be working.		* Then submit an '''array of 2-3 jobs''', to make sure that you fully understand how array will be working.
	* Only then, commit to submitting the '''full data set''' processing as the complete array.		* Only then, commit to submitting the '''full data set''' processing as the complete array.


	If you submit an array of thousands of jobs and all of them will fail, this will cause a significant strain on the cluster.		If you submit an array of thousands of jobs and all of them will fail, this will cause a significant strain on the cluster.

Dmitri: /* Array jobs */

2024-11-28T22:31:25Z

Array jobs

← Older revision		Revision as of 22:31, 28 November 2024
Line 1:		Line 1:
			= Warning =
			Job arrays is an '''advanced technique'''. Some '''caution should be used''' when submitting array of jobs.

			Please never submit a full set of jobs when starting a new set of calculation.
			* Please, '''first time''' submit '''only one job of the set''' as a test case and make sure that it runs and completes successfully.
			* Then submit an '''array of 2-3 jobs''', to make sure that you fully understand how array will be working.
			* Only then, commit to submitting the '''full data set''' processing as the complete array.

			If you submit an array of thousands of jobs and all of them will fail, this will cause a significant strain on the cluster.
			The scheduler will schedule all these jobs and will be preparing nodes for them and then cleaning up the nodes after the failed jobs.
			This will be a lot of work done for nothing.
			Please be considerate.

	= Array jobs =		= Array jobs =

Lleung: Added navbox

2023-09-21T20:36:23Z

Added navbox

← Older revision		Revision as of 20:36, 21 September 2023
Line 168:		Line 168:
	[[Category:Guides]]		[[Category:Guides]]
	[[Category:How-Tos]]		[[Category:How-Tos]]
			{{Navbox Guides}}

Lleung: Added categories

2023-09-20T21:57:17Z

Added categories

← Older revision		Revision as of 21:57, 20 September 2023
Line 165:		Line 165:
	SLURM_ARRAY_TASK_MIN=1		SLURM_ARRAY_TASK_MIN=1
	</pre>		</pre>

			[[Category:Guides]]
			[[Category:How-Tos]]

Dmitri: /* Array jobs */

2023-01-30T16:59:15Z

Array jobs

← Older revision		Revision as of 16:59, 30 January 2023
Line 43:		Line 43:

	With this approach, it is left to the <code>myapplication</code> to interpret what the input parameter,		With this approach, it is left to the <code>myapplication</code> to interpret what the input parameter,
	in this case <code>1</code> or <code>8</code>, mean.		in this case <code>1</code> or <code>8</code>, mean, in terms of data processing.

	= Mapping task index to a data file =		= Mapping task index to a data file =

Dmitri: /* Array jobs */

2023-01-30T16:58:34Z

Array jobs

← Older revision		Revision as of 16:58, 30 January 2023
Line 43:		Line 43:

	With this approach, it is left to the <code>myapplication</code> to interpret what the input parameter,		With this approach, it is left to the <code>myapplication</code> to interpret what the input parameter,
	in this case <code>1</code> or <code>8</code> mean.		in this case <code>1</code> or <code>8</code>, mean.

	= Mapping task index to a data file =		= Mapping task index to a data file =

Dmitri: /* Array jobs */

2023-01-30T16:58:06Z

Array jobs

← Older revision		Revision as of 16:58, 30 January 2023
Line 41:		Line 41:
	The resources requested in the job script are allocated to each elemental job separately,		The resources requested in the job script are allocated to each elemental job separately,
	that is, each job gets its own 4 GB of RAM and 1 CPU, as requested in the example.		that is, each job gets its own 4 GB of RAM and 1 CPU, as requested in the example.

			With this approach, it is left to the <code>myapplication</code> to interpret what the input parameter,
			in this case <code>1</code> or <code>8</code> mean.

	= Mapping task index to a data file =		= Mapping task index to a data file =

Dmitri: /* Array options */

2022-08-05T17:37:37Z

Array options

← Older revision		Revision as of 17:37, 5 August 2022
Line 137:		Line 137:
	having the values of <code>{1, 5, 9, .... 97}</code>.		having the values of <code>{1, 5, 9, .... 97}</code>.

	Another option is to indicate the number of element jobs that can run at the same time.		Another option is to indicate the '''number of element jobs that can run at the same time'''.
	By default all the element jobs can be executed at the same time, provided that resources are available.		By default all the element jobs can be executed at the same time, provided that resources are available.
	With this option you can limit the number of jobs to, for example, one:		With this option you can limit the number of jobs to, for example, one:
Line 144:		Line 144:
	you can combine it with the stride as well, to limit your 25 jobs to run 5 at a time:		you can combine it with the stride as well, to limit your 25 jobs to run 5 at a time:
	#SBATCH --array=1-100:4%5		#SBATCH --array=1-100:4%5

			Limiting a job array to run 1 job at a time may be a practical way to '''resubmit the same job'''
			a fixed number of times.
			If you want to run some simulation for 1 000 000 steps, it can be done as an array job of, say 100 element,
			each element running 10 000 steps, one after another.

	== Array Environmental variables ==		== Array Environmental variables ==

Dmitri: /* Mapping task index to a data file */

2022-08-05T17:34:32Z

Mapping task index to a data file

← Older revision		Revision as of 17:34, 5 August 2022
Line 126:		Line 126:
	# ====================================================		# ====================================================
	</syntaxhighlight>		</syntaxhighlight>

			= More Arrays =

			== Array options ==

			You can indicate a '''stride for the array elements''' to be used in element jobs:
			#SBATCH --array=1-100:4

			This option in your job script sets the stride to 4 and will run 25 jobs with <code>SLURM_ARRAY_TASK_ID</code>
			having the values of <code>{1, 5, 9, .... 97}</code>.

			Another option is to indicate the number of element jobs that can run at the same time.
			By default all the element jobs can be executed at the same time, provided that resources are available.
			With this option you can limit the number of jobs to, for example, one:
			#SBATCH --array=1-100%1

			you can combine it with the stride as well, to limit your 25 jobs to run 5 at a time:
			#SBATCH --array=1-100:4%5

			== Array Environmental variables ==

			For a first element jobs in an <code>--array=1-100</code>
			the following variables will be set to
			<pre>
			SLURM_JOB_ID=87654321
			SLURM_ARRAY_JOB_ID=87654321
			SLURM_ARRAY_TASK_ID=1
			SLURM_ARRAY_TASK_COUNT=100
			SLURM_ARRAY_TASK_MAX=100
			SLURM_ARRAY_TASK_MIN=1
			</pre>