On Slurm on Atos AD complex was updated to version 22.05. Since AD has been the default cluster with hpc-loginand hpc-batchbeing aliases for nodes on AD complex.
The same version of Slurm 22.05 was has also been installed on AA and AB complexes and will be installed on AC complex on .
One change in the new Slurm version impacts all batch jobs setting the number of OMP threads with directive:
Code Block
languagetheme
bashConfluence
#SBATCH --cpus-per-task
According to Following recommendation from the official release note:
Panel
borderColor
black
borderStyle
solid
title
Slurm 22.05release note:
Srun will no longer read in SLURM_CPUS_PER_TASK. This means you will implicitly have to specify --cpus-per-task on your srun calls, or set the new SRUN_CPUS_PER_TASK env var to accomplish the same thing.
Users should adjust the jobs script by exporting SRUN_CPUS_PER_TASK environmental variable manually:
or alternatively we strongly suggest to all Atos HPC users adjusting their job scripts by specifying number of OMP threads as an option with run "srun" command:
Code Block
language
bash
srun --cpus-per-task
To reduce user impact and made job scripts compatible with the new Slurm, HPC support team has set SRUN_CPUS_PER_TASK environmental variable :