Skip to main content
Skip to main menu Skip to spotlight region Skip to secondary region Skip to UGA region Skip to Tertiary region Skip to Quaternary region Skip to unit footer

Slideshow

Sample Slurm Submit Scripts

Sample Script

This first script uses the SLURM_JOBID variable to create a directory to store the results of the job



#!/bin/bash

#SBATCH --job-name=mclz             # Job name

#SBATCH --partition=batch           # Partition (queue) name

#SBATCH --nodes=1                   # Number of nodes

#SBATCH --ntasks-per-node=1         # Number of tasks to call on each node

#SBATCH --mem-per-cpu=200mb         # Memory per processor

#SBATCH --time=24:00:00             # Time limit hrs:min:sec

#SBATCH --output=mclz.%j.out        # Standard output log

#SBATCH --error=mclz.%j.err         # Standard error log
JOBDIR=${SLURM_JOBID}

mkdir $JOBDIR
export PATH=./bin:$PATH

time srun ./bin/mclz.sh -n -I C -Z 6 -A He -L TA -N "TA"
cp c6+he* $JOBDIR

Sample OpenMPI Script

#!/bin/sh

#SBATCH --nodes=5                      # Number of nodes

#SBATCH --ntasks-per-node=6            # Number of tasks per node

#SBATCH --mem=20                       # Memory per CPU (MB)

#SBATCH -p batch                       # Partition to use

#SBATCH --output=example.%j.out        # Standard output log

#SBATCH --error=example.%j.err         # Standard error log
export PMIX_MCA_psec=native

export OMPI_MCA_btl="tcp,self"

srun centmpi
 

Sample Intel MPI Script



#!/bin/bash

#SBATCH --job-name=water0

#SBATCH --partition=batch             # Partition (queue) name

#SBATCH --nodes=1                     # Number of nodes

#SBATCH --ntasks=2                    # Number of MPI ranks

#SBATCH --ntasks-per-node=2           # Number of tasks to run on each node

#SBATCH --cpus-per-task=1             # Number of cores per MPI rank

#SBATCH --ntasks-per-core=1           # Number of tasks to call on each core

#SBATCH --output=water0-%j.out        # Standard output log

#SBATCH --error=water0-%j.err         # Standard error log
export PMIX_MCA_psec=native           # Required for munge to work

#export OMPI_MCA_btl="tcp,self"        # Required for OpenMPI
## If compiling before submitting, run these three commands before compiling.

## Otherwise, run these three commands before submitting your job, or remove the '###' to uncomment them

###module purge

###module load intel/2022.1.0

###module load impi/2021.6.0
unset I_MPI_PMI_LIBRARY               # When using impi and mpirun
mpirun -np 2 ./water0_MPI.exe
 

Sample OpenMP/Intel MPI Hybrid Script



#!/bin/bash

#SBATCH --job-name=test1

#SBATCH --partition=batch             # Partition (queue) name

#SBATCH --nodes=1                     # Number of nodes

#SBATCH --ntasks=4                    # Number of MPI ranks

#SBATCH --ntasks-per-node=4           # Number of tasks to run on each node

#SBATCH --output=test1-%j.out        # Standard output log

#SBATCH --error=test1-%j.err         # Standard error log
# Since this code is compiled with intel mpi compilers, swap out gcc/OpenMPI

# You can also run these three module commands from the command line before submitting your job

# Comment them out here if you run these commands from the command line before job submission.

module purge

module load intel/2022.1.0

module load impi/2021.6.0

 

Support us

We appreciate your financial support. Your gift is important to us and helps support critical opportunities for students and faculty alike, including lectures, travel support, and any number of educational events that augment the classroom experience. Click here to learn more about how to help us grow.

Every dollar given has a direct impact upon our students and faculty.