How to use Ansys-Fluent

Software	Version	Cluster
Ansys-Fluent	21.2	Dardel

Submitting a Fluent job on Dardel

Before preparing your script, you need to make a journal file to give tasks to Fluent. It should be in the same directory as your job subbmission script.

# -----------------------------------------------------------
# SAMPLE JOURNAL FILE
#
# read case file (*.cas) that had previously been prepared file/read-case fluent.cas
# read data file (*.dat) in case you are starting simulation from a saved file
# comment it if you start from zero
file/read-data mycase.dat
# run 10 iterations
solve/iterate 10
# write data
file/write-data results.dat
# exit fluent
exit yes
# ------------------------------------------------------------

Then you can use the following script for running Ansys/Fluent on Dardel:

#!/bin/bash 

# Name of your job
#SBATCH -J test

#SBATCH -A <allocation>

# Wall-clock time will be given to this job
#SBATCH -t 10:00:00

# Number of nodes
#SBATCH -N 1

# Set the partition for your job. 
#SBATCH -p <partition>

#SBATCH -e error_file.e%J

# load module fluent v16.0
module load PDC
module load fluent/21.2 

# The Journal file
JOURNALFILE=journalfile.jou

# Set the license server below
export ANSYSLMD_LICENSE_FILE=<license file>
export ANSYSLI_SERVERS=<license server>
# If you are KTH user, contact PDC support or KTH IT for license server information
# export some necessary environment variables
export LC_ALL=C; unset LANGUAGE
# Total number of Processors
# NPROCS=1x128 = number of nodes x number of processors per node
NPROCS=128

fluent 3ddp -g -t $NPROCS -mpi=cray -i $JOURNALFILE > my_output_file.$SLURM_JOBID

Then you can submit your job using sbatch

sbatch fluent_run.sh

Disclaimer

PDC takes no responsibility for the correctness of results produced with the binaries. Always evaluate the binaries against known results for the systems and properties you are investigating before using the binaries for production jobs.