Abaqus

Abaqus is an application that is used for solving structural simulation of multi-physics problems.

There are two main modes of running Abaqus on Octopus:

• batch mode (using the command line or scripts)

• graphical user interface mode (using desktop sessions)

To use any of these two the abaqus module should be loaded:

module load abaqus

Graphical user interface mode

To launch the Abaqus GUI after connecting to a desktop environment, to launch CAE (Complete Abaqus Environment) the following command can be executed in a terminal:

module load abaqus
abaqus cae

Template Abaqus job (batch mode)

The following job script can be used as a template to run abaqus batch jobs on one compute node.

#!/bin/bash

## specify the job and project name
#SBATCH --job-name=abaqus
#SBATCH --account=ab123

## specify the required resources
#SBATCH --partition=normal
#SBATCH --nodes=1
#SBATCH --ntasks-per-node=1
#SBATCH --cpus-per-task=16
#SBATCH --mem=64000
#SBATCH --time=1-00:00:00

source ~/.bashrc
module load abaqus/2020
abaqus job=my_abaqus_sim_name input=my_sim.inp cpus=`nproc` mp_mode=threads interactive

Multi-node parallel Abaqus jobs

The following job script below can be used to run a parallel Abaqus job using multiple compute nodes in batch mode. The script below can be downloaded by clicking here

If a graphica user interface job is used then the following script should be executed in the same folder where the command abaqus cae -mesa is executed:

module load abaqus
chmod +x slurm_abaqus_mpi_env_gen.sh
./slurm_abaqus_mpi_env_gen.sh
abaqus cae -mesa

After the job is executed MPI must be selected in the the ABAQUS job in the GUI.

script for running a multi-node parallel job in batch mode

#!/bin/bash

## specify the job and project name
#SBATCH --job-name=abaqus
#SBATCH --account=ab123

## specify the required resources
#SBATCH --partition=large
#SBATCH --nodes=4
#SBATCH --ntasks-per-node=1
#SBATCH --cpus-per-task=64
#SBATCH --mem=64000
#SBATCH --time=1-00:00:00

source ~/.bashrc
module load abaqus/2020

##############################################################
# DO NOT MODIFY BEYOND THIS UNLESS YOU KNOW WHAT YOU ARE DOING
##############################################################

# dump the hosts to a text file
SLURM_HOSTS_FILE=slurm-hosts-${SLURM_JOBID}.out

#
# generate the mp_host_list environment variable
#
srun hostname > ${SLURM_HOSTS_FILE}

mp_host_list="["
for HOST in `sort ${SLURM_HOSTS_FILE} | uniq`; do
    echo ${HOST}
    mp_host_list="${mp_host_list}""['${HOST}',`grep ${HOST} ${SLURM_HOSTS_FILE} | wc -l`]"
done

mp_host_list=`echo ${mp_host_list} | sed 's/\]\[/\]\,\[/g'`"]"

echo $mp_host_list

#
# write the abaqus environment file
#
ABAQUS_ENV_FILE="abaqus_v6.env"
cat > ${ABAQUS_ENV_FILE} << EOF
import os
os.environ['ABA_BATCH_OVERRIDE'] = '1'
verbose=3
mp_host_list=${mp_host_list}
if 'SLURM_PROCID' in os.environ:
    del os.environ['SLURM_PROCID']
EOF

abaqus job=my_input_file.inp cpus=$SLURM_NPROCS` -verbose 3 standard_parallel=all mp_mode=mpi interactive