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 125: | Line 125: | ||
<!--T:19--> | <!--T:19--> | ||
Requesting <code>--mem=0</code> is interpreted by Slurm to mean | Requesting <code>--mem=0</code> is interpreted by Slurm to mean <i>reserve all the available memory on each node assigned to the job.</i> | ||
<!--T:30--> | <!--T:30--> | ||
If you need more memory per node than the smallest node provides (e.g. more than 125 GiB at Graham) then you should not use <code>--mem=0</code>, but request the amount explicitly. Furthermore, some memory on each node is reserved for the operating system. To find the largest amount your job can request and still qualify for a given node type, consult the | If you need more memory per node than the smallest node provides (e.g. more than 125 GiB at Graham) then you should not use <code>--mem=0</code>, but request the amount explicitly. Furthermore, some memory on each node is reserved for the operating system. To find the largest amount your job can request and still qualify for a given node type, consult the <i>Available memory</i> column of the <i>Node characteristics</i> table on cluster description page: | ||
* [[Béluga/en#Node_characteristics|Béluga node characteristics]] | * [[Béluga/en#Node_characteristics|Béluga node characteristics]] | ||
* [[Cedar#Node_characteristics|Cedar node characteristics]] | * [[Cedar#Node_characteristics|Cedar node characteristics]] | ||