Pitzer
Planned Globus maintenance downtime will take place on Saturday, October 24, 2026, from 10:30 AM to 11:30 AM EDT (9:30 AM US Central). Globus services will undergo database upgrades and will be unavailable for up to 1 hour, affecting Globus Auth, Transfer, Groups, Flows, Search, Timers, Compute, GlobusID, and Action Providers. Check Globus page for more details: https://www.globus.org/blog/2026-10-24-maintenance.
HDF5 2.2.0 available on Pitzer
Date:
Wednesday, October 14, 2026 - 3:45pm
System(s):
HDF5 2.2.0 is available on Pitzer with gcc 12.3.0 and either OpenMPI 5.0.10 or MVAPICH 4.1
use module spider hdf5/2.2.0 to see more
Scratch Space Purge Schedule Update: Changing to Daily Removals
Due to ongoing scratch space challenges and high utilization, the data purge frequency is changing from every Wednesday to daily. Files and data that have not been accessed for 60 days or more will be automatically removed from the system every day. Please ensure that you regularly back up or move any important data to your project or permanent storage directories to prevent unintended loss.
Emergency Scratch Purge on August 26
Please be advised that an emergency scratch purge will be executed on OSC Research Data Storage today, August 26, 2026, starting from 3pm. See this for more details: https://www.osc.edu/resources/technical_support/known_issues/emergency_maintenance_research_data_storage_scratch_purge
Expanding Resource Selection Across All OnDemand Apps
Following our recent update to the OnDemand (OOD) Desktop launch form, we are expanding this job type resource interface across all applications within the OOD instance throug
System Downtime August 11 2026
A downtime for OSC HPC systems is scheduled from 7 a.m. to 9 p.m., Tuesday, August 11, 2026. The downtime will affect the Pitzer, Cardinal and Ascend Clusters, web portals, the state-wide licenses and HPC file servers. MyOSC (the client portal) will be available during the downtime. In preparation for the downtime, the batch scheduler will not start jobs that cannot be completed before 7 a.m., August 11. Jobs that are not started on clusters will be held until after the downtime and then started once the system is returned to production status.