Native Run

Ok, now we’ll run the mpi binary natively.

cat > native-hello-run.sbatch << EOF
#!/bin/bash
#SBATCH --job-name=native-hello-run
#SBATCH --ntasks=4
#SBATCH --output=/fsx/logs/%x_%j.out

module load intelmpi
unset I_MPI_PMI_LIBRARY
set -x
mpirun /fsx/bin/hello
EOF

We’ll submit the job with two nodes.

sbatch -N2 native-hello-run.sbatch

Depending on what you did already on the cluster, it might take a couple of minutes to spin up the two compute nodes.

squeue

cat /fsx/logs/native-hello-run_*.out