Updates


Event Date Summary
The upgrade and expansion was completed Mar. 22, 23:30 (Pacific)

Incident description

System Incident status Start Date End Date
Cedar Closed
Created by Martin Siegert on

Title


Scheduled Cedar Upgrade and Expansion


Summary


The Cedar system will be unavailable during March 16-22, 2020 due to a major system expansion and upgrade. During this week-long downtime, the Cedar team will be adding 768 compute nodes (36864 cores) and 192 GPU nodes (768 GPU cards), as well as: - upgrading BIOS and firmware of compute nodes and network switches, - upgrading the operating system, - upgrading the file system software and storage controller firmware, - upgrading the network drivers, - modifying the scheduling software, - benchmarking the new system. Cedar will be unavailable starting at 6am (Pacific time) on March 16 and is expected to be available again on March 23. This site will receive updates during the downtime in case the status changes. All jobs that are still running at the start of the downtime will be terminated and need to be resubmitted once the system is back online.


Updated by Pier-Luc St-Onge on