October 27th chip and RRStor Downtime Completed
To all chip users,
The downtime for the chip cluster scheduled for October 27th has been completed, and the cluster is now back online and accessible. We appreciate your patience while we upgraded our core infrastructure for chip.
Please note that some cluster nodes are still booting after the downtime. We expect these to all be back online within the next few hours.
Notable changes
We’ve made some changes to the slurm backfill schedulers to better handle the job queue when there are many jobs (>500) pending. Users should expect their smaller jobs to be scheduled to start sooner if there are idle nodes in the cluster.
We have also enabled the scrontab feature within slurm to allow users to run periodic jobs on the cluster. We will follow up shortly with details on how to use this feature.
If you encounter any issues, please submit a support ticket here: https://rtforms.umbc.edu/rt_authenticated/doit/DoIT-support.php?auto=Research%20Computing
For additional information, you can check our documentation at: https://umbc.atlassian.net/wiki/spaces/faq/pages/1082589207/UMBC+HPCF+-+chip
Thank you,
Gregory Ballantine
HPC Specialist
Posted: October 27, 2026, 6:02 PM