The Daily Bulletin

Cheyenne system administrators will perform preventative maintenance procedures on the PBS job scheduler Thursday, May 24, beginning at noon MDT. The maintenance is expected to take less than one hour to complete.

Most PBS commands will not work during the maintenance outage, including qstat and qsub, and new job submissions will not be possible.

No batch jobs are expected to be lost as a result of the maintenance. Jobs that are executing when the maintenance begins will continue to run without interruption. Jobs that are queued for execution or in a hold state will remain in those states until PBS is returned to service. Access to Cheyenne’s login nodes will not be interrupted.

Users will be notified when PBS is returned to service.

Cheyenne users and others who transfer files via the Globus Command Line Interface (CLI) are advised to make sure they are using the CLI application, which is installed on the CISL data-access nodes. The CLI application has replaced the hosted legacy CLI, which Globus will no longer support starting August 1.

The CISL Globus file transfers documentation describes how to get started using the CLI application and includes links to detailed Globus documentation. Some user scripts that are configured to make unattended transfers may need to be revised.

NCAR’s new Casper cluster, which is expected to go into production this summer, is a heterogeneous system of specialized data analysis and visualization resources and large-memory, multi-GPU nodes. Named for the city in Wyoming, Casper is the successor to Geyser and Caldera, which have been in service since 2012 and will be decommissioned by the end of 2018.

The new system, procured from PCPC Direct, Ltd., consists of a total of 24 Supermicro nodes featuring Intel Skylake processors. See the new Casper web page for additional details, and look for more information in the CISL Daily Bulletin in the coming weeks.

A new and larger work space is now available for GLADE. The new /glade2/work file space has a total capacity of 2 PB. The default quota for individual users 1 TB. (Users can access the new space from the Cheyenne login nodes as /glade/work but the symlinks are not yet active on the Cheyenne batch nodes.)

The older /glade/p/work file space (where users’ default quota is 512 GB) will become read-only within the next two months. The date will be announced soon. The /glade/p/work space will be retired by the end of 2018.

Users are encouraged to begin using the new space soon and move all of their active files and directories from the old space to their individual /glade/work directories at their earliest convenience.

Based on input from users, the new /glade/work spaces will remain unpurged. We apologize for the incorrect info in yesterday's Daily B item.

CISL is asking for input from researchers in order to gauge their interest in Phase 6 of the Coupled Model Intercomparison Project (CMIP6) and planned use of CISL’s CMIP Analysis Platform.

NCAR data products are expected in mid-2018 and will be available to CMIP Analysis Platform users on the GLADE file system. As estimated time frames for availability of data from other modeling centers become known, CISL will publish that information for the community to help researchers make their plans. When other data sets are available, researchers who use the CMIP Analysis Platform will be able to submit requests to have the data sets they need added to GLADE.

Researchers who are or may be planning to use CMIP6 data are asked to respond to this survey by Friday, May 25, and are encouraged to share the survey with others.

The CISL User Services Section will present an in-person and online tutorial at 9:30 a.m. Friday, May 25, at the NCAR Mesa Lab in Boulder for new users of NCAR’s high-performance computing (HPC) systems. The 60-minute tutorial will cover basic usage and typical workflows for Cheyenne, Geyser, and Caldera.

Topics include:

  • The Cheyenne computing environment

  • Accessing software, including compilers and MPI libraries

  • Submitting batch jobs using the PBS scheduler

  • Using Slurm to schedule data analysis and visualization tasks

  • A quick look ahead to the new Casper system

Register to attend in person—at the VisLab in NCAR’s Mesa Lab (ML4) in Boulder—or via webcast by selecting a link below: