Scholar Cluster Maintenance
As of 1:39pm EST, engineers have completed maintenance and have returned the Scholar cluster back to normal service. All queues have been enabled and job scheduling have resumed.
There have been some important updates to the cluster. Please make sure to review detailed list of changes to verify that your necessary applications continue to run in the new software and hardware environment. Please report any issues to firstname.lastname@example.org
Link to original posting ORIGINAL:
The Scholar cluster will be unavailable December 21, 2022 8:00am - December 22, 2022 5:00pm EST for scheduled maintenance between semesters. The cluster will return to full production by Thursday, December 22nd, 2022 at 5:00pm EST.
During this time, Scholar will have hardware life-time update and replacement, the operating system patched, and its user-facing applications and libraries stack updated. Most important changes ate outlined below:
- New modern AMD EPYC 3rd generation CPU nodes, boosting Scholar CPU capacity from 400 cores to 512 cores;
- Up to date version of CUDA libraries for the GPU nodes;
- Up to date version of the Open OnDemand gateway with better job management and interactive applications capabilities;
- Updated stack of CPU- and GPU-based applications;
- Improved management of
- Accounts audit to ensure compliance with Scholar accounts policy;
How does this affect you?
- You will be unable to login to Scholar or access data stored there during the maintenance.
- Jobs requesting a walltime which would cause the job to run beyond the start of maintenance will not start.
- All pending and running jobs will be deleted during the maintenance, as there will be a hardware change on affected partitions.
- After maintenance you will need to ensure your
module loadcalls and your compiled application continue to work in the new software environment. In some cases you may have to adjust module versions and/or recompile your application.
Your access to Scholar may be affected!
- Students: If your access to Scholar was lingering from a class you have taken in the past and happened to not been properly removed after the end of that class, you may find your access removed. Please make sure you have copied out any valuable data ahead of time!
- Instructors: to ensure your Scholar account stays between semesters, please make sure you register your future courses with us as soon as possible.
Please reach out to email@example.com if you have any questions about this maintenance or need support. Please note that RCAC staff will observe the university winter recess from December 23, 2022 through January 2, 2023.