PACE A Partnership for an Advanced Computing Environment

Dear Researchers,

Last week, we completed the migration of our third batch of users from Rich to Coda datacenter, which was a major milestone for our research community as we migrated nearly 2,000 active users from clusters in Rich to the Phoenix cluster in Coda.  Given the more than doubled number of users on already highly utilized Phoenix cluster coupled with the grace period that we have in effect for job accounting with respect to recent announcement about the new cost model, we have noticed a rapid increase in wait time per job by users on the cluster.   At this time, we are updating our scheduler policy to alleviate pressure on the wait time per job for users that should improve the overall quality of service.  The changes listed below are data-driven and have been carefully chosen so as to not adversely impact research teams that submit large scale jobs.

Effective today, the following changes have been made to the scheduler policy that effect the inferno and embers queues:

• Reduced the concurrent-use limit for CPU usage per research group from 7,200 processors to 6,000 processors.
• Reduced the concurrent-use limit for  GPU usage per user from 220 GPUs to 32 GPUs
• Added a per research group concurrent CPU hour capacity limit set to 300,000 CPU hours that allows the scheduler to permit the research group to concurrently run up to 300,000 CPU hours (i.e., requested processors * walltime)
• Added a per job CPU-time capacity limit set to 264,960 CPU hours that would allow, for example, a 2,208 core job to run for 5 days.

Jobs that violate these limits will be held in the queue until currently running jobs complete and the total number of utilized processors, GPUs, and/or the remaining CPU-time fall below the thresholds.   We have updated our documentation to reflect these changes, which you may view here.

Again, the changes listed above are taking effect today, December 14, 2020.   If you have any questions or concerns, please direct them to pace-support@oit.gatech.edu.

All the best,

The PACE Team

[Update – December 9, 2020 – 11:03am]

Thank you for your continued patience.   We are reaching out to update you that the scheduler has been resolved late last night, new jobs from users that were held in queue yesterday have resumed, and we have observed normal operation of the Phoenix scheduler through the night and this morning.  The cause of the scheduler issue was the large sudden influx of user jobs on December 7 as previously reported.  The updated timeout parameter will be kept in place to prevent a similar occurrence in the future.   Additionally, we are expanding our alert utilities to track additional metrics from the scheduler to alert us of similar stressors so that we may proactively address/mitigate scheduler issues.

The Phoenix cluster is ready for research.  Thank you again for your patience as we worked to address this issue in coordination with the vendor.

[Update – December 8, 2020 – 6:31pm]

[Original Note – December 8, 2020 – 10:26am]

Dear PACE users,

PACE is investigating a scheduler issue that is impacting the Phoenix cluster.  At this time, users are unable to run jobs, and  jobs are held in queue.  

This is an active situation, and we will follow up with updates as they become available.

Thank you for your attention to this urgent message, and we apologize for this inconvenience.