Allocations and compute scheduling: Difference between revisions

Jump to navigation Jump to search
Give GPU-equivalents in MB to better reflect how memory resources are managed
(Add Beluga bundle characteristics)
(Give GPU-equivalents in MB to better reflect how memory resources are managed)
Line 89: Line 89:
<!--T:29-->
<!--T:29-->
Compute Canada systems have the following GPU-core-memory bundle characteristics:
Compute Canada systems have the following GPU-core-memory bundle characteristics:
* Cedar: 1 GPU / 6 cores / 32GB
* Cedar: 1 GPU / 6 cores / 32000 MB
* Graham: 1 GPU / 8 cores / 32GB
* Graham: 1 GPU / 8 cores / 32000 MB
* Béluga: 1 GPU / 10 cores / 47GB
* Béluga: 1 GPU / 10 cores / 47000 MB


<!--T:30-->
<!--T:30-->
[[Category:SLURM]]
[[Category:SLURM]]
</translate>
</translate>
cc_staff
353

edits

Navigation menu