We need to shut down all DGX-B200 nodes for urgent maintenance.
- New jobs and currently pending jobs will not start on these nodes
- Running jobs will keep going for now, but we'll kill any jobs still running at 6 am tomorrow (July 7th 2026)
- If you expect your jobs to run beyond 6am, please take steps to ensure that you can start from an appropriate checkpoint later.
- The rest of the cluster is unaffected by this maintenance.
Run this command to identify your affected jobs
squeue --node=gpu[4001-4008] -u $USER
Please reach out to us at support@ccv.brown.edu if you have any questions. Sincerely, CCV Research Technical Services Team