The Daily Bulletin

The video and slides from CISL’s January 12 class introducing users to the new Cheyenne supercomputer are now available here in the CISL Course Library.

This 70-minute tutorial covers typical workflows in the Cheyenne environment as well as performance and operational differences between the Yellowstone and Cheyenne systems.

Related user documentation links include:

The new 5.34-petaflops Cheyenne HPC system has been released for production work by Accelerated Scientific Discovery (ASD) projects. CISL expects to make the system available to the general user community within the next couple of weeks.

CISL staff are monitoring the new system closely and providing assistance as needed for the ASD users, whose large-scale university and NCAR projects have priority during Cheyenne’s first 10 weeks in production. Some software packages and modules are still being installed and tested.

Introduction of the Cheyenne system, which was built for NCAR by SGI, is accompanied by expansion of the integrated GLADE shared-disk resource. The GLADE system’s total usable capacity increased from 16 to 31 PB with recent additions and will grow to 36 PB in the next few months. The new components transfer data at a rate of 200 GPps, more than twice as fast as the older file systems.

Users who have questions are encouraged to contact the CISL consultant on duty or cislhelp@ucar.edu. Links to user documentation can be found here: Cheyenne.

Ensemble runs, data assimilation runs, and other jobs on Yellowstone can generate tens or hundreds of thousands of output files, log files, and others over time. Such large numbers of files can be difficult to manage and remove from GLADE file spaces when they are no longer needed. Configuring jobs to place no more than 2,000 to 3,000 files in a single directory will make them easier to manage.

See Removing large numbers of files for how to remove massive accumulations of files and see this page for more best practices.