Using GPUs with Slurm/fr: Difference between revisions

Using GPUs with Slurm/fr (view source)

3 bytes removed , 6 years ago

no edit summary

rsnt_translations

56,437

edits

@@ Line 102: / Line 102: @@
 === Ordonnancement d'un nœud ''GPU large'' pour Cedar ===
-<div class="mw-translate-fuzzy">
+Sur la grappe [[Cedar/fr|Cedar]] se trouve des nœuds de type ''GPU large'' qui ont chacun quatre cartes Tesla P100 16Go; les autres nœuds sont de 12Go. Les GPUs du groupe spécial utilisent le all use the same PCI switch so the inter-GPU communication latency is lower, but bandwidth between CPU and GPU is lower than on the regular GPU nodes. The nodes also have 256 GB RAM instead of 128GB. In order to use these nodes you must specify <code>lgpu</code>. By-gpu requests can '''only run up to 24 hours'''.
-There is a special group of large-memory GPU nodes at [[Cedar]] which have four Tesla P100 16GB cards each. (Other GPUs in the cluster have 12GB.) These GPUs all use the same PCI switch so the inter-GPU communication latency is lower, but bandwidth between CPU and GPU is lower than on the regular GPU nodes. The nodes also have 256 GB RAM instead of 128GB. In order to use these nodes you must specify <code>lgpu</code>. By-gpu requests can '''only run up to 24 hours'''.
-</div>
 {{File