The Daily Bulletin delivers news about CISL’s high-performance computing environment, training opportunities, scheduled maintenance, and other events of interest to the user community.
Downtime is scheduled for Quasar Tuesday, January 25th at 9:00 am MT, expected until approximately 10:00 am. None planned for Cheyenne, Casper, Glade, Campaign Storage, Stratus, or JupyterHub.
Some Cheyenne user documentation has been updated to provide more detail about the importance of specifying an ompthreads value in parallel job scripts even if they do not use OpenMP threads. See Cheyenne job script examples for more information about when this is particularly important.
None planned for Cheyenne, Casper, Glade, Campaign Storage, Quasar, Stratus, or JupyterHub.
A new submission queue called "gpudev" is now available on the Casper cluster to support GPU application development and debugging efforts. It provides rapid access to up to four V100 GPUs, avoiding the sometimes lengthy queue wait times in the "gpgpu" execution queue. Job submissions to the new queue are limited to 30 minutes walltime instead of the 24-hour limit for all other Casper submissi...
CISL HPC engineers will perform maintenance on three of the Casper cluster’s GPU nodes this morning beginning at 9 a.m. MST. Three of the system’s five 8xV100 nodes (each of which hosts eight NVIDIA V100 GPUs) will be unavailable for approximately three hours. The other two 8xV100 nodes will remain available, as will other Casper GPU nodes identified in this system documentation.
To help the NCAR and university research communities prepare to use the extensive GPU-computing capabilities of the new Derecho system, CISL is launching a series of workshops and tutorials on Thursday, February 3, at 10 a.m. MST for scientists, software engineers, and students in the Earth system sciences. The series of one-hour sessions will be held every two weeks through August using the Zo...