Seitenvergleich

...

If no specific requests regarding the number of tasks has been done, mpirun defaults to hyperthreading and starts cores*2 processes. If a number of tasks has been specified (either with -nN or --tasks-per-node), mpirun  honors this via the flag -map-by. For example:

#!/bin/bash
#SBATCH -N 4

module load gcc/9.3.0 openmpi/gcc.9/4.1.4

cpus_per_node=${SLURM_JOB_CPUS_PER_NODE%\(*}

mpirun -map-by ppr:$cpus_per_node:node ./yourexe

...