NCAR’s HPC resources will be unavailable to users May 9-13 while CISL staff perform annual system and facility maintenance at the NCAR-Wyoming Supercomputing Center (NWSC). The work includes a number of HPC system software updates in addition to power and cooling system maintenance tasks. Progress will be communicated through Notifier emails during the week.
General operating system, security updates, and driver and firmware updates are planned for Cheyenne and Casper; the GLADE, Quasar, and Stratus storage resources; and the HPC/storage systems network. These updates are expected to have minimal user-level impact, and CISL staff will verify resource functionality before releasing the systems to the user community. JupyterHub will also be upgraded at this time, with numerous enhancements that are available now on the development branch. Users can evaluate the development JupyterHub instance for a preview of these features and assess the functionality for their own workflows.
Scheduler reservations are in place now on both Cheyenne and Casper to ensure that all user jobs have completed by 6 a.m. MDT on May 9 as the downtime begins. This also means that jobs requesting long walltimes will not start as the deadline approaches and the queues drain. Any jobs that are queued when the downtime begins will be retained for execution when the systems return to service.