August HPCF Newsletter

Happy end of August everyone!

I'm writing to give everyone an update regarding the UMBC HPCF and other Research Computing projects currently undertaken by DoIT during the month of August.

Summary

• Preemption Notifications
• Downtime Scheduled for September 29th
• Slurm Queueing Updates
• SIG Committee Meetings Scheduled

Preemption Notifications

As mentioned in the post-downtime message from August 25, there will now be a notification message written to the Slurm error file of a preempted job. If a job is preempted, the associated Slurm error file will see a message like “[SYSTEM NOTIFICATION] Job XYZ was PREEMPTED by Slurm.” along with a timestamp and other information.

We will add more reports such as this that will help users figure out what resources were used by their jobs.

Downtime Scheduled for September 29th

The DoIT Research Computing & Data team has scheduled a downtime on September 29th. Please plan your workloads and batch jobs accordingly, as cluster services will be temporarily unavailable during this window and jobs will be canceled during this window. We appreciate your patience as we complete these updates to ensure a more robust and high-performing computing environment.

We will send out a reminder with additional details two weeks prior to the downtime.

Slurm Queueing Updates

Earlier in the month, many users noted that job wait time increased due to a large volume of job submissions. Minor changes were made to Slurm to better react to cases such as these, effectively limiting the number of jobs that can accrue priority that are submitted by a single user. This change and other topics will be discussed at both the CPU and GPU SIG sub-committee meetings in early September.

If you experience unexpectedly long wait times on your submitted slurm jobs, please submit a descriptive RT Ticket so we can investigate!

Shared Infrastructure Governance Committees Scheduled

The various sub-committees of the Shared Infrastructure Governance Steering Committee have been scheduled. These sub-committees each work to address strategy, new ideas, and issues affecting the research computing cyberinfrastructure of the campus. Meeting locations will be announced via the HPCF myUMBC Group. Please see the Shared Infrastructure Governance webpage for more information

• Advanced CPU Infrastructure sub-committee meetings will take place on the second Friday of each month starting at 1500ET.
• Advanced GPU Infrastructure sub-committee meetings will take place on the first Monday of each month starting at noon ET. 
• Cloud Infrastructure sub-committee meetings will take place on the second Wednesday of each month, starting at noon ET. 
• User Experience sub-committee meetings will take place on the second Monday of each month starting at noon ET. 

Publications
If you have any publications, presentations, theses, or other works that made use of the campus cluster, please submit an RT Ticket with bibliographic information so that we can accurately reflect this work in our records and on the HPCF Website. 

Need Help?
As always, please communicate any issues/questions to the Research Computing RT Queue (hpcf.umbc.edu > User Support > Request Help).

Gregory Ballantine,
HPC System Administrator