Additional nodes in the Casper cluster are now accessible using the PBS Pro workload manager as the migration from Slurm to PBS continues. In addition to the 64 new high-throughput computing (HTC) nodes announced recently, about half of the cluster’s other nodes have been transitioned as of today. PBS jobs can now run on:
- 7 CPU-only nodes
- 4 GP100 GPU nodes
- 2 4xV100 GPU nodes
- 3 8xV100 GPU nodes
The rest of the Casper CPU-only and GPU nodes will move from Slurm to PBS on Wednesday, April 7. At that point all Casper jobs will be submitted via PBS. This new and updated documentation provides details for preparing PBS scripts and submitting jobs on Casper:
Please
contact CISL if you have questions or need assistance.