BATCH Scripts in ARINA

Our aim is to help researchers:

We prepared several BATCH scripts available in:

/home/users/slurm/batch-scripts/

SOFTWARE
ABINIT	BLAST	generic	Lumerical
nf_core_eager	qespresso	scilab	starccm+
ams	CP2K	gromacs	matlab
nwchem	qiime2	shasta	Trinity
ANSYS	lammps	molcas
orca	R	siesta	vasp

---

Example SLURM Script (slurm.sl)

The following example shows a general BATCH script that can be modified or adapted to most of the common calculations ran in ARINA:

#!/bin/bash 
#SBATCH --ntasks=4
#SBATCH --mem-per-cpu=3760M
#SBATCH --time=4:00:00
#SBATCH --partition=vfast

# Required input files for the software execution,
# separated by a space
files="input.inp checkpoint.chk"

# Unload all possible present modules
module purge
# LOAD NOW REQUIRED MODULES
module load MYDESIREDMODULE/version-compiler

# Define the command line you will run
command=""

#######################################################
## Do not change below this line

. /home/users/slurm/etc/slurm_func.sh

# Signal trap to recover files
trap 'cleanup_function' EXIT

HOST=$(hostname)
echo "Job running on $HOST"

chooseScr

LDIR=$(pwd)
export LDIR

echo "cp ${files} $0 to $scr/."
cp ${files} $0 ${scr}/.

cd $scr

# CALL EXECUTABLE with SRUN
echo /usr/bin/time -p srun ${command}
eval /usr/bin/time -p srun ${command}

Once the script is generated or modified, the sbatch command has to be executed in order to send the job to the cluster queues.

Submitting a job

[user@agamede] sbatch slurm.sl

---

Notes

• The scripts are designed for ARINA's SLURM-managed clusters.
• slurm_func.sh contains helper functions for scratch space and file recovery.
• chooseScr accepts the -g option to enforce to use the /gscratch/$USER directory.
• Always edit command and files variables according to the specific software you are running.
• Modules should be loaded as needed before execution.