Helix/Slurm - Revision history

H Schumacher: Added info about --begin paramer

2026-02-27T00:29:09Z

Added info about --begin paramer

← Older revision		Revision as of 02:29, 27 February 2026
Line 209:		Line 209:
	Once the walltime limit has been reached you will be automatically logged out from the compute node.		Once the walltime limit has been reached you will be automatically logged out from the compute node.

	For convenient access to specific GUI applications (JupyterLab, ...) on the cluster, we provide a web-based platform: [[Helix/bwVisu \| bwVisu]]		For convenient access to specific GUI applications (JupyterLab, RStudio, ...) on the cluster, we provide a web-based platform: '''[[Helix/bwVisu \| bwVisu]]'''

			'''Delay the start of a job''' by using [https://slurm.schedmd.com/sbatch.html#OPT_begin -b] (or [https://slurm.schedmd.com/sbatch.html#OPT_begin --begin]) option in order to defer the allocation of the job until the specified time. Examples:
			<syntaxhighlight lang="bash">$ salloc --begin=20:00 ... # job can start after 8 p.m.
			$ salloc --begin=now+1hour ... # job can start 1 hour after submission
			$ salloc --begin=teatime ... # job can start at teatime (4 p.m.)
			$ salloc --begin=2023-12-24T20:00:00 ... # job can start after specified date/time
			</syntaxhighlight>

	= Job Monitoring =		= Job Monitoring =
Line 217:		Line 224:
	For an overview of your submitted jobs use the command:		For an overview of your submitted jobs use the command:

	<syntaxhighlight lang="bash">$ squeue~~</syntaxhighlight>~~		<syntaxhighlight lang="bash">$ squeue
			# Additionally show the estimated start time:
			$ squeue --start</syntaxhighlight>

	To get detailed information about a specific job use the command:		To get detailed information about a specific job use the command:

H Winkhardt: /* GPU Programs */

2025-12-12T14:50:23Z

GPU Programs

← Older revision		Revision as of 16:50, 12 December 2025
Line 174:		Line 174:
	#SBATCH --gres=gpu:A40:1		#SBATCH --gres=gpu:A40:1
	#SBATCH --time=12:00:00		#SBATCH --time=12:00:00
	#SBATCH --mem=~~59gb~~		#SBATCH --mem=16gb

	module load devel/cuda		module load devel/cuda

H Winkhardt: GPU example script update for Helix

2025-12-12T14:49:53Z

GPU example script update for Helix

← Older revision		Revision as of 16:49, 12 December 2025
Line 168:		Line 168:
	<syntaxhighlight lang="slurm">		<syntaxhighlight lang="slurm">
	#!/bin/bash		#!/bin/bash
	#SBATCH --partition=gpu-single		#SBATCH --partition=gpu-single
	#SBATCH --nodes=1		#SBATCH --nodes=1
	#SBATCH --ntasks~~-per-node~~=40		#SBATCH --ntasks=1
	#SBATCH --~~gres~~=~~gpu:A40:4~~		#SBATCH --cpus-per-task=8
			#SBATCH --gres=gpu:A40:1
	#SBATCH --time=12:00:00		#SBATCH --time=12:00:00
	#SBATCH --mem=~~200gb~~		#SBATCH --mem=59gb

	module load devel/cuda		module load devel/cuda
	export OMP_NUM_THREADS=${~~SLURM_NTASKS~~}		export OMP_NUM_THREADS=${SLURM_CPUS_PER_TASK}
	./my_cuda_program		./my_cuda_program
	</syntaxhighlight>		</syntaxhighlight>

H Winkhardt at 08:47, 17 June 2025

2025-06-17T08:47:41Z

← Older revision		Revision as of 10:47, 17 June 2025
Line 12:		Line 12:
	\| [https://slurm.schedmd.com/sbatch.html sbatch] \|\| Submits a job and queues it in an input queue		\| [https://slurm.schedmd.com/sbatch.html sbatch] \|\| Submits a job and queues it in an input queue
	\|-		\|-
	\| [https://slurm.schedmd.com/salloc.html ~~saclloc~~] \|\| Request resources for an interactive job		\| [https://slurm.schedmd.com/salloc.html salloc] \|\| Request resources for an interactive job
	\|-		\|-
	\| [https://slurm.schedmd.com/squeue.html squeue] \|\| Displays information about active, eligible, blocked, and/or recently completed jobs		\| [https://slurm.schedmd.com/squeue.html squeue] \|\| Displays information about active, eligible, blocked, and/or recently completed jobs

H Schumacher: small correction

2025-05-21T15:05:08Z

small correction

← Older revision		Revision as of 17:05, 21 May 2025
Line 95:		Line 95:
	* The number of GPUs is requested with the option "--gres=gpu:<number-of-gpus>".		* The number of GPUs is requested with the option "--gres=gpu:<number-of-gpus>".
	* A specific GPU type can be requested with the option "--gres=gpu:<gpu-type>:<number-of-gpus>". Possible values for <gpu-type> are listed in the line 'GPU Type' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].		* A specific GPU type can be requested with the option "--gres=gpu:<gpu-type>:<number-of-gpus>". Possible values for <gpu-type> are listed in the line 'GPU Type' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].
	* GPUs that are suitable for a specific GPU memory requirement can be requested with option "--gres=gpu:<number-of-gpus>,gpumem_per_gpu:<required-gpumem>G". This only restricts the selection of possible GPU types. For the job the total GPU memory per GPU is available as listed in the line 'GPU memory per GPU' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].		* GPUs that are suitable for a specific GPU memory requirement can be requested with option "--gres=gpu:<number-of-gpus>,gpumem_per_gpu:<required-gpumem>GB". This only restricts the selection of possible GPU types. For the job the total GPU memory per GPU is available as listed in the line 'GPU memory per GPU' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].

	== Constraints ==		== Constraints ==

H Schumacher: small correction for gpumem_per_gpu

2025-05-21T08:54:05Z

small correction for gpumem_per_gpu

← Older revision		Revision as of 10:54, 21 May 2025
Line 95:		Line 95:
	* The number of GPUs is requested with the option "--gres=gpu:<number-of-gpus>".		* The number of GPUs is requested with the option "--gres=gpu:<number-of-gpus>".
	* A specific GPU type can be requested with the option "--gres=gpu:<gpu-type>:<number-of-gpus>". Possible values for <gpu-type> are listed in the line 'GPU Type' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].		* A specific GPU type can be requested with the option "--gres=gpu:<gpu-type>:<number-of-gpus>". Possible values for <gpu-type> are listed in the line 'GPU Type' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].
	* GPUs that are suitable for a specific GPU memory requirement can be requested with option "--gres=gpu:<number-of-gpus>,gpumem_per_gpu:<required-gpumem>GB". This only restricts the selection of possible GPU types. For the job the total GPU memory per GPU is available as listed in the line 'GPU memory per GPU' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].		* GPUs that are suitable for a specific GPU memory requirement can be requested with option "--gres=gpu:<number-of-gpus>,gpumem_per_gpu:<required-gpumem>G". This only restricts the selection of possible GPU types. For the job the total GPU memory per GPU is available as listed in the line 'GPU memory per GPU' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].

	== Constraints ==		== Constraints ==

S Richling: /* GPU Programs */

2025-04-24T09:37:29Z

GPU Programs

← Older revision		Revision as of 11:37, 24 April 2025
Line 187:		Line 187:
	** A40 GPUs are optimized for single precision computations.		** A40 GPUs are optimized for single precision computations.
	** A100 and H200 GPUs offer better performance for double precision computations or if the code makes use of tensor cores.		** A100 and H200 GPUs offer better performance for double precision computations or if the code makes use of tensor cores.
	* The ~~Cuda~~ module used for compilation must be loaded before the start of the program.		* The CUDA module used for compilation must be loaded before the start of the program.

	=== More examples ===		=== More examples ===

S Richling: /* GPU Programs */

2025-04-24T09:34:06Z

GPU Programs

← Older revision		Revision as of 11:34, 24 April 2025
Line 171:		Line 171:
	#SBATCH --nodes=1		#SBATCH --nodes=1
	#SBATCH --ntasks-per-node=40		#SBATCH --ntasks-per-node=40
	#SBATCH --gres=gpu:4		#SBATCH --gres=gpu:A40:4
	#SBATCH --time=12:00:00		#SBATCH --time=12:00:00
	#SBATCH --mem=200gb		#SBATCH --mem=200gb

S Richling: /* GPU Programs */

2025-04-24T09:32:27Z

GPU Programs

← Older revision		Revision as of 11:32, 24 April 2025
Line 184:		Line 184:
	** Example for a request of two A40 GPUs: --gres=gpu:A40:2		** Example for a request of two A40 GPUs: --gres=gpu:A40:2
	** Example for a request of one A100 GPU: --gres=gpu:A100:1		** Example for a request of one A100 GPU: --gres=gpu:A100:1
	* If you are unsure, if your code ~~will run~~ faster ~~on a A40 or a A100~~, please run a test case and compare the run times. In general the following applies:		* If you are unsure on which GPU type your code runs faster, please run a test case and compare the run times. In general the following applies:
	** A40 GPUs are optimized for single precision computations.		** A40 GPUs are optimized for single precision computations.
	** A100 GPUs offer better performance for double precision computations or if the code makes use of tensor cores.		** A100 and H200 GPUs offer better performance for double precision computations or if the code makes use of tensor cores.
	* The Cuda module used for compilation must be loaded before the start of the program.		* The Cuda module used for compilation must be loaded before the start of the program.

S Richling: /* GPU requests */

2025-04-24T09:20:30Z

GPU requests

← Older revision		Revision as of 11:20, 24 April 2025
Line 95:		Line 95:
	* The number of GPUs is requested with the option "--gres=gpu:<number-of-gpus>".		* The number of GPUs is requested with the option "--gres=gpu:<number-of-gpus>".
	* A specific GPU type can be requested with the option "--gres=gpu:<gpu-type>:<number-of-gpus>". Possible values for <gpu-type> are listed in the line 'GPU Type' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].		* A specific GPU type can be requested with the option "--gres=gpu:<gpu-type>:<number-of-gpus>". Possible values for <gpu-type> are listed in the line 'GPU Type' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].
	* ~~GPU types~~ that are suitable for a specific GPU memory requirement can be requested with option "--gres=gpu:<number-of-gpus>,gpumem_per_gpu:<required-gpumem>GB". This only restricts the selection of possible GPU types. For the job the total GPU memory per GPU is available as listed in the line 'GPU memory per GPU' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].		* GPUs that are suitable for a specific GPU memory requirement can be requested with option "--gres=gpu:<number-of-gpus>,gpumem_per_gpu:<required-gpumem>GB". This only restricts the selection of possible GPU types. For the job the total GPU memory per GPU is available as listed in the line 'GPU memory per GPU' of the [https://wiki.bwhpc.de/e/Helix/Hardware#Compute_Nodes Compute Nodes table].

	== Constraints ==		== Constraints ==