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. To get the Daily Bulletin by email, use this link to subscribe.
The CISL Consulting Services Group will present an online tutorial at 9 a.m. MDT on Tuesday, March 30, to help Casper users transition from the Slurm scheduler to PBS Pro. Users who have experience with PBS on Cheyenne are encouraged to attend as well, as there are a few differences between the PBS deployments on each system. Topics will include: • Batch and interactive scheduling with qsub, qs...
You’ll see a fresh new look soon when you check your Daily Bulletin email, which is the first piece of the brand new portal being developed for our HPC user community. The NCAR Advanced Research Computing portal will be rolled out in just a few days as the new home of the Daily Bulletin, a quick way to check on HPC system status, and a link to the NCAR Research Computing help center. That’s jus...
HPSS will be down March 23 from 7am to 4pm for hardware work on two of the libraries. No downtime scheduled for Cheyenne, Casper, GLADE, Campaign Storage, JupyterHub, Quasar, or Stratus.
Casper users will notice some service changes as the cluster is migrated from Slurm to the PBS job scheduler over the next couple of weeks. The first change involves the vncmgr script, which allows users to start VNC sessions on Casper nodes. This script uses Slurm now, but will be migrated to PBS on Monday, March 29. Changes will be minimal, but custom scheduler options must be specified in PB...
The NCAR/UCAR virtual Python Tutorial Seminar Series continues on Wednesday with an introduction to the Python package Matplotlib at 1 p.m. MDT. See the Xdev team’s blog for details. Participants who did not attend the earlier Jupyter Notebooks tutorial are advised to watch this recording before attending the Matplotlib session.
The Casper cluster now has 64 new high-throughput computing (HTC) nodes, which are for batch jobs that require the use of one full node or less. As the first step in migrating the Casper system’s scheduler from Slurm to the PBS Pro workload manager, the HTC nodes are accessible only through PBS Pro. This new documentation provides details for preparing PBS scripts and submitting jobs on Casper:...