...
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 -n or with -N and/or --tasks-per-node), an (Open) mpirun honors this via the flag -map-by. For example:
| Codeblock |
|---|
| language | bash |
|---|
| title | Hyperthreading active, 80/192 Processes per Node will be started |
|---|
| collapse | true |
|---|
|
#!/bin/bash
#SBATCH -N 4
module load gcc/9.23.0 openmpi/gcc.9/34.1.54
cpus_per_node=${SLURM_JOB_CPUS_PER_NODE%\(*}
mpirun -map-ppnby cpusppr:$cpus_per_node:node ./yourexe |
| Codeblock |
|---|
| language | bash |
|---|
| title | Hyperthreading disabled, 96 processes per node are started. |
|---|
| collapse | true |
|---|
|
#!/bin/bash
#SBATCH -N 4
#SBATCH --tasks-per-node 96
module load gcc/9.23.0 openmpi/gcc.9/34.1.54
tasks_per_node=${SLURM_TASKS_PER_NODE%\(*}
mpirun -ppn tasksmap-by ppr:$tasks_per_node:node ./yourexe |