PACE A Partnership for an Advanced Computing Environment

[Update – November 3, 2018, 4:45pm] 

Our November 2018 maintenance is complete on schedule. We have brought compute nodes online and released previously submitted jobs. Login nodes are accessible and your data are available. As usual, there are a small number of straggling nodes we will address over the coming days, which includes nodes that will need PCIe connectors replaced as a preventative measure.

Please note that some jobs on PACE ICE cluster need to be re-submitted, and we have contacted the affected users individually. 

[Update – November 2, 2018] 

Verification of the integrity of the GPFS file system is taking longer than initial estimate.  As a result, this maintenance day will last the full three days, as scheduled, that will allow us to complete the certification of the file system and ensure the highest integrity of the data. 

[Original Post – October 29, 2018]

Our next PACE maintenance day will start November 1 and run through November 3 as scheduled.

As usual, jobs with long walltimes will be held by the scheduler to ensure that no active jobs will be running when systems are powered off. These jobs will be released as soon as the maintenance activities are complete. You can reduce the walltime of such jobs to ensure completion before 6am on November 1 and resubmit if this will give them enough time to complete successfully.

Planned Tasks

Compute

• (no user action needed) Replace power components in a rack in Rich 133
• (no user action needed) Replace defective PCIe connectors on multiple servers

Network

• (no user action needed) Stress test new InfiniBand subnet managers, to prepare for the move to Coda
• (no user action needed) Change uplink connections from management switches

Storage

• (no user action needed) Verify integrity of GPFS file systems
• (no user action needed) Upgrade firmware on DDN / GPFS storage systems
• (no user action needed) Upgrade firmware on TruNAS storage systems

Other

• (no user action needed) Replace PACE ICE schedulers with a physical server, to increase capacity and reliability

[Update – October 29, 2018] Abaqus and Ansys license servers are restored.

[Original Post – October 26, 2018] On Thursday, October 25, multiple virtual servers experienced problems due to data corruption from a storage issue.  OIT’s  storage team is working to correct this matter and operations is working on rebuilding the machines so they can be restored.  This service interruption has impacted the license servers for Ansys and Abaqus.  This has impacted PACE users’ Ansys and Abaqus jobs.  If you had Ansys and Abaqus jobs submitted during this period, please check your jobs, and resubmit them once license server is back online.

Currently, Ansys has been brought back online.

For additional information regarding this incident, please follow the status page link at https://status.gatech.edu

We experienced a failure in the primary InfiniBand subnet manager that may have impacted both running and starting jobs.   The malfunction happened in such a way that the backup IB subnet manager (SM) didn’t notice the primary was failing to operate normally. We disabled the primary SM, and the secondary SM took over as designed. The service outage lasted from 12:56pm to 01:07pm today, October 15, 2018. PACE staff will continue to investigate this failure mode and adjust the procedures to help prevent it in the future.  As this brief network interruption may have impacted the running and starting jobs, please check your jobs to identify if there are any crashed jobs and report any problems you may notice to pace-support@oit.gatech.edu

[Update – October 5, 2018] We worked with our vendor to address the issue impacting the network shared disk (NSD) that drastically reduced the performance of pace1 file system when it was stressed by the large number of I/O intensive jobs.  On Thursday, we had the NSD restored to normal, and our benchmarks indicate a successful resolution.  As a precaution, we will continue to monitor NSDs as the user workloads continue to resume to normal.

[Original Post – October 3, 2018]

On Monday, October 1, we started to experience slowness on our parallel file system (pace1), which was associated with users’ I/O intensive jobs.  We have engaged the users who were/are responsible for the load.  During this process, the stress on our storage and network allowed us to identify a bug with a network shared disk that is responsible for caching data that improves read/write speeds.  Currently, we have successfully deployed a workaround, which has dramatically improved the performance, and we are working with our vendor to further resolve this issue.

With this development, symptoms that you may have experienced is slowness when navigating through your files.  Your jobs should not have been impacted other than slower access to the files that may have resulted in longer execution times (i.e., wall-time).

We will update you once we have the issue fully resolved in collaboration with our vendor.  If you have any questions, please don’t hesitate to contact us at pace-support@oit.gatech.edu