rsnt_translations
56,420
edits
Advanced MPI scheduling: Difference between revisions
Jump to navigation
Jump to search
no edit summary
No edit summary |
No edit summary |
||
| Line 169: | Line 169: | ||
--mem=96G | --mem=96G | ||
srun --cpus-per-task=$SLURM_CPUS_PER_TASK application.exe | srun --cpus-per-task=$SLURM_CPUS_PER_TASK application.exe | ||
This job is the same size as the last one: 16 tasks (that is, 16 MPI processes), each with 4 threads. The difference here is that we are sure of getting exactly 2 whole nodes. | This job is the same size as the last one: 16 tasks (that is, 16 MPI processes), each with 4 threads. The difference here is that we are sure of getting exactly 2 whole nodes. Remember that <code>--mem</code> requests memory <i>per node</i>, so we use it instead of <code>--mem-per-cpu</code> for the reason described earlier. | ||
=== Why srun instead of mpiexec or mpirun? === <!--T:10--> | === Why srun instead of mpiexec or mpirun? === <!--T:10--> | ||