ANSYS - Revision history

Dmitri: /* ANSYS Fluent on ARC */

2025-07-10T17:31:14Z

ANSYS Fluent on ARC

← Older revision		Revision as of 17:31, 10 July 2025
Line 40:		Line 40:

	'''Ansys versions''':		'''Ansys versions''':
	* <code>ansys/2024r1</code> -- supported. Works. <code>-mpi=~~intel -ic=ib~~</code>		* <code>ansys/2024r1</code> -- supported. Works. <code>-mpi=openmpi</code>
	* <code>ansys/2021r1</code> -- Does not work due to technical problems.		* <code>ansys/2021r1</code> -- Does not work due to technical problems.
	* <code>ansys/2020r2</code> -- supported. Works. <code>-mpi=openmpi -ic=eth</code>		* <code>ansys/2020r2</code> -- supported. Works. <code>-mpi=openmpi -ic=eth</code>

Dmitri: /* Example Job Scripts */

2025-03-03T17:10:14Z

Example Job Scripts

← Older revision		Revision as of 17:10, 3 March 2025
Line 155:		Line 155:
	# -----------------------------------------------------------------------------------------------		# -----------------------------------------------------------------------------------------------
	INPUT=elbow3.in		INPUT=elbow3.in
	fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=~~intel -pib~~ -i $INPUT		fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=openmpi -i $INPUT
	</syntaxhighlight>		</syntaxhighlight>

Line 180:		Line 180:
	# -----------------------------------------------------------------------------------------------		# -----------------------------------------------------------------------------------------------
	INPUT=elbow3.in		INPUT=elbow3.in
	fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=~~intel -pib~~ -i $INPUT		fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=openmpi -i $INPUT
	</syntaxhighlight>		</syntaxhighlight>

Line 205:		Line 205:
	# -----------------------------------------------------------------------------------------------		# -----------------------------------------------------------------------------------------------
	INPUT=elbow3.in		INPUT=elbow3.in
	fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=~~intel -pib~~ -i $INPUT		fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=openmpi -i $INPUT
	</syntaxhighlight>		</syntaxhighlight>

Dmitri: /* Introduction */

2025-01-20T18:27:49Z

Introduction

← Older revision		Revision as of 18:27, 20 January 2025
Line 11:		Line 11:
	The software can be downloaded, upon approval, from the [https://iac01.ucalgary.ca/SDSWeb/ IT Software Distribution] web site.		The software can be downloaded, upon approval, from the [https://iac01.ucalgary.ca/SDSWeb/ IT Software Distribution] web site.

	ANSYS is available to all U of C researchers with an ARC account, but, with licensing restrictions as outlined in the next section.= ~~<span id="licensing"></span>~~Licensing considerations =		ANSYS is available to all U of C researchers with an ARC account, but, with licensing restrictions as outlined in the next section.

			=Licensing considerations =

	For many years, Information Technologies has provided a limited number of license tokens for ANSYS software, sometimes supplemented by contributions from researchers. The software contract is typically renewed annually in August. If you are interested in contributing to the pool of licenses, you can write to the IT Help Desk [mailto:itsupport@ucalgary.ca itsupport@ucalgary.ca] and ask that your email be redirected to the IT software librarian.		For many years, Information Technologies has provided a limited number of license tokens for ANSYS software, sometimes supplemented by contributions from researchers. The software contract is typically renewed annually in August. If you are interested in contributing to the pool of licenses, you can write to the IT Help Desk [mailto:itsupport@ucalgary.ca itsupport@ucalgary.ca] and ask that your email be redirected to the IT software librarian.

Dmitri: /* Fluent Timings */

2024-03-18T17:35:23Z

Fluent Timings

← Older revision		Revision as of 17:35, 18 March 2024
Line 210:		Line 210:
	== Fluent Timings ==		== Fluent Timings ==

	These timings are provided for reference purpose.		These timings are provided for '''reference purpose'''.
	They have been obtained on some example system that runs for about 10 minutes on 2 full compute nodes of the '''cpu2019''' partition.		They have been obtained on some example system that runs for about 10 minutes on 2 full compute nodes of the '''cpu2019''' partition.

			=== 2024r1 timings ===

	<pre>		<pre>
	Partition #Nodes #CPUs/Procs ~~MPI Interconnect~~ Walltime Fluent		Partition #Nodes #CPUs/Procs Walltime Fluent
	---------------------------------------------------------------------------------------------------		---------------------------------------------------------------------------------------------------
	cpu2019 2 ~~80 openmpi vendor eth 10~~:36 ~~2020r2~~		cpu2019 1 40 15:00 2024r1
			cpu2019 2 80 9:13 2024r1
	~~openmpi local eth 7~~:49 ~~2020r2~~		cpu2019 3 120 7:22 2024r1
	~~openmpi local ib~~ 7:~~53 2020r2~~		cpu2019 4 160 6.29 2024r1

	~~openmpi local ib 8:01~~ ~~2019r2~~

	~~ibmmpi vendor eth 8:34 2019r2~~
	~~ibmmpi vendor ib~~.~~infinipath 7:56~~ ~~2019r2~~

			cpu2021 1 48 13:10 2024r1
			cpu2021 2 96 8:10 2024r1
			cpu2021 3 144 6:35 2024r1

	~~parallel~~ ~~8 96 openmpi vendor eth 19~~:12 ~~2020r2~~		cpu2022 1 52 10:56 2024r1
	~~openmpi vendor ib 13~~:25 ~~2020r2~~		cpu2022 2 104 6:56 2024r1
			cpu2022 3 156 5:57 2024r1

	~~openmpi local eth 12~~:18 ~~2020r2~~		cpu2023 1 64 10:25 2024r1
	~~openmpi local ib 12~~:25 ~~2020r2~~		cpu2023 2 128 7:18 2024r1
			cpu2023 3 192 6:36 2024r1
	</pre>		</pre>

Dmitri: /* Job script for cpu2019 partition */

2024-03-18T17:24:25Z

Job script for cpu2019 partition

← Older revision		Revision as of 17:24, 18 March 2024
Line 140:		Line 140:
	#SBATCH --ntasks-per-node=40		#SBATCH --ntasks-per-node=40
	#SBATCH --cpus-per-task=1		#SBATCH --cpus-per-task=1
	#SBATCH --mem=~~160GB~~		#SBATCH --mem=128GB
	#SBATCH --time=01:00:00		#SBATCH --time=01:00:00
	#SBATCH --partition=cpu2019,cpu2019-bf05		#SBATCH --partition=cpu2019,cpu2019-bf05

Dmitri: /* Fluent Timings */

2024-03-18T17:15:48Z

Fluent Timings

← Older revision		Revision as of 17:15, 18 March 2024
Line 205:		Line 205:
	fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=intel -pib -i $INPUT		fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=intel -pib -i $INPUT
	</syntaxhighlight>		</syntaxhighlight>

			= Timings and Benchmarks =

	== Fluent Timings ==		== Fluent Timings ==

Dmitri: /* Example Job Scripts */

2024-03-18T17:15:10Z

Example Job Scripts

← Older revision		Revision as of 17:15, 18 March 2024
Line 122:		Line 122:

	== Example Job Scripts ==		== Example Job Scripts ==

			The best '''ARC's partitions''' to run ANSYS Fluent are '''cpu2019''', '''cpu2021''', and '''cpu2022'''.
			They have '''40''', '''48''', and '''52''' '''CPUs per node'''.

			When running jobs on one of those partitions, the jobs should be requesting '''all the CPUs on the requested nodes'''.

			Below are '''example job scripts''' to run '''Fluent 2024r1''' on these partitions.
			These script can be accordingly changed to use '''older versions of Fluent''' installed on ARC by changing the '''ANSYS module'''.

			=== Job script for <code>cpu2019</code> partition ===

	<code>fluent2024r1_cpu2019.slurm</code>:		<code>fluent2024r1_cpu2019.slurm</code>:
Line 146:		Line 156:
	</syntaxhighlight>		</syntaxhighlight>

			=== Job script for <code>cpu2021</code> partition ===

	<code>fluent2024r1_cpu2021.slurm</code>:		<code>fluent2024r1_cpu2021.slurm</code>:
Line 169:		Line 180:
	fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=intel -pib -i $INPUT		fluent 3ddp -g -t${SLURM_NTASKS} -cnf=${HOSTLIST} -mpi=intel -pib -i $INPUT
	</syntaxhighlight>		</syntaxhighlight>

			=== Job script for <code>cpu2022</code> partition ===

	<code>fluent2024r1_cpu2022.slurm</code>:		<code>fluent2024r1_cpu2022.slurm</code>:

Dmitri: /* Example Job Scripts */

2024-03-18T17:08:38Z

Example Job Scripts

@@ Line 135: / Line 135: @@
 # -----------------------------------------------------------------------------------------------
-module load ansys/2024r2
+module load ansys/2024r1
 # Create a node list so that Fluent knows which nodes to use.

Dmitri: /* Example Job Scripts */

2024-03-18T17:06:54Z

Example Job Scripts

← Older revision		Revision as of 17:06, 18 March 2024
Line 123:		Line 123:
	== Example Job Scripts ==		== Example Job Scripts ==

	<code>~~fluent2020r2_cpu2019~~.slurm</code>:		<code>fluent2024r1_cpu2019.slurm</code>:
	<syntaxhighlight lang="bash">		<syntaxhighlight lang="bash">
	#!/bin/bash		#!/bin/bash
Line 129:		Line 129:
	#SBATCH --nodes=1		#SBATCH --nodes=1
	#SBATCH --ntasks-per-node=40		#SBATCH --ntasks-per-node=40
			#SBATCH --cpus-per-task=1
	#SBATCH --mem=160GB		#SBATCH --mem=160GB
	#SBATCH --time=01:00:00		#SBATCH --time=01:00:00
Line 134:		Line 135:

	# -----------------------------------------------------------------------------------------------		# -----------------------------------------------------------------------------------------------
	module load ansys/~~2020r2~~		module load ansys/2024r2
	~~module load openmpi/4.1.1-gnu~~
	~~export OPENMPI_ROOT=$(dirname $(dirname `which mpiexec`))~~

	# Create a node list so that Fluent knows which nodes to use.		# Create a node list so that Fluent knows which nodes to use.

Dmitri: /* ANSYS Fluent on ARC */

2024-03-15T21:57:31Z

ANSYS Fluent on ARC

← Older revision		Revision as of 21:57, 15 March 2024
Line 38:		Line 38:

	'''Ansys versions''':		'''Ansys versions''':
	* <code>ansys/2024r1</code> -- supported. Works.		* <code>ansys/2024r1</code> -- supported. Works. <code>-mpi=intel -ic=ib</code>
	* <code>ansys/2021r1</code> -- Does not work due to technical problems.		* <code>ansys/2021r1</code> -- Does not work due to technical problems.
	* <code>ansys/2020r2</code> -- supported. Works.		* <code>ansys/2020r2</code> -- supported. Works. <code>-mpi=openmpi -ic=eth</code>
	* <code>ansys/2019r2</code> -- supported. Works.		* <code>ansys/2019r2</code> -- supported. Works. <code>-mpi=openmpi -ic=eth</code>

	== Creating a Fluent input file ==		== Creating a Fluent input file ==