NCAR Advanced Research Computing

The new home for news, information, and support for CISL's HPC users. More features coming!

Extended HPC systems downtime scheduled for July 26 through August 2

May 11, 2021

All HPC compute and data storage systems hosted at the NCAR-Wyoming Supercomputing Center will be unavailable during part or all of a planned maintenance downtime beginning Monday, July 26, to allow CISL staff and contractors to perform critical updates to the facility’s power infrastructure in preparation for the arrival of the new Derecho system. The Casper cluster, GLADE file systems, Campai...

No downtime scheduled for May 10-14

May 10, 2021

No downtime scheduled for Cheyenne, Casper, GLADE, JupyterHub, Quasar, Stratus, or HPSS

Q. Where should I run small jobs that only use CPUs?

May 6, 2021

A. The Casper cluster’s high-throughput computing (HTC) nodes. Why? Take a look at the specs. You’ll see that Casper has 64 HTC nodes specifically for running small batch jobs. It also has: • More shared resources than the Cheyenne share queue • A far higher concurrent-use limit for CPU cores than Cheyenne: 468 vs. 36. • More available memory, plus NVMe swap for overflow. Here’s an example sc...

Postal Address: P.O. Box 3000, Boulder, CO 80307-3000 • Shipping Address: 3090 Center Green Drive, Boulder, CO 80301