Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 11 Current »

This how-to article explains and demonstrates how to launch the Ansys fluent suite on HPC.

\uD83D\uDCD8 Instructions

Module File

First check the available fluent modules

module avail fluent

To load fluent version 2023R1:

module load fluent/2023R1

Parallel Job

To run several tasks in parallel on one or more nodes, here is an example fluentP.sh batch script for older versions of Fluent:

#!/bin/bash
#SBATCH -N 2 # allocate 2 nodes for the job
#SBATCH --ntasks-per-node=20
##SBATCH --exclusive # no other jobs on the nodes while job is running
#SBATCH -o fluentP_%J.out # the file to write stdout for fluent job
#SBATCH -e fluentP_%J.err # the file to write stderr for fluent job

source /etc/profile.d/modules.sh
module load fluent/$VERSION

if [ -z "$SLURM_NPROCS" ]; then
  N=$(( $(echo $SLURM_TASKS_PER_NODE | sed -r 's/([0-9]+)\(x([0-9]+)\)/\1 * \2/') ))
else
  N=$SLURM_NPROCS
fi

echo -e "N: $N\n";

# run fluent in batch on the allocated node(s)
fluent 2ddp -g -slurm -t$N -pdefault (or -pinfiniband) -ssh -i foo.txt

Make sure to replace foo.txt in the last line with the name of your job file. Then, submit your job with:

sbatch fluentP.sh

Parallel Job for new fluent versions

Here is an example fluentP.sh batch script for newer versions of Fluent:

#!/bin/bash
#SBATCH -N 2 # allocate 2 nodes for the job
#SBATCH --ntasks-per-node=20
#SBATCH --exclusive # no other jobs on the nodes while job is running
#SBATCH -o fluentP_%J.out # the file to write stdout for fluent job
#SBATCH -e fluentP_%J.err # the file to write stderr for fluent job
module purge # allow for a clean module environment for the submission
module load gcc/11.3.0 zlib/1.2.12 java/1.8.0_162 mpi/openmpi/3.1.3 fluent/2019R3 # load the most current fluent module to be used along with other modules
# run fluent in batch mode on the allocated node(s) using either one of the types of fluent solvers 2d,2ddp,3d, or 3ddp
# The -g option is used to run without GUI or graphics, The -slurm option is used to specify the scheduler, -t is used to specify the number of processors "cores" for the fluent model to use
# The -i option is used to specify the input file for fluent to execute, the file would need to be in the SAME directory where this submission script is located
# The standard error and output will be generated with the above format in the SBATCH headers.  The names for these files can change to the name of the specific project currently being worked 
on.
# the fluent -h or fluent -help commands will list available options for fluent that can be used
fluent 3ddp -g -slurm -t$SLURM_NTASKS -pdefault -i foo.txt

Make sure to replace foo.txt in the last line with the name of your job input file. Then, submit your job with:

sbatch fluentP.sh

Interactive Job

Start the interactive session on 1 node with 12 CPUs with fisbatch:

fisbatch -N 1 -n 12

Fluent

The first parameter to fluent is the version (namely, 2ddp below). The version can be either 2d, 3d, 2ddp, or 3ddp; change it to whatever you need. The "dp" stands for "double precision".

Run fluent with:

fluent 2ddp -g -t $SLURM_NTASKS

Type "exit" to end the session.

Ansys Electromagnetics Suite

Run the graphical launcher for Ansys Electromagnetics Suite with:

ansysedt

OSX users should not use XQuartz 2.7.9 because of an OpenGL bug and should instead downgrade to version 2.7.8

User Guide

The command-line help usage actually provides very limited information:

fluent -help

The only place Ansys supplies to read more documentation is from the "Help" inside the Fluent GUI. Open a terminal in X2Go, load the fluent module, and then run `fluent` from the command-line to launch the GUI. Click on the Help button > Help on Starting and Executing Fluent.

  • No labels