GPU Nodes Available for Testing in CentOS 7 Environment
GPU nodes are now available for testing in the CentOS 7 environment.
We have 5 nodes available in the pascal partition and 2 nodes available in the maxwell partition, so for now, please limit your jobs to short tests to verify that your application functions correctly. Please continue running your production jobs in the CentOS 6 environment.
We would like to move over more GPU resources within the next 2-3 weeks, and ideally have all GPU nodes moved over within the next 6-8 weeks, so please test as soon as possible.
Important note: In order to improve and automate account administration, we have changed the names of GPU groups (SLURM accounts) in the CentOS 7 environment. You can easily check the name of your GPU account by running the “slurm_groups” command while logged into the CentOS 7 environment.
Let us know if you have any questions or need assistance rebuilding your application to run in the CentOS 7 environment by opening a helpdesk ticket with us.