The Daily Bulletin

Downtime Monday, August 29: Yellowstone, Geyser_Caldera, Glade 12:00 - 1:00

No Scheduled Downtime: HPSS


NCAR researchers and computational scientists are encouraged to submit requests for NCAR Strategic Capability (NSC) projects to be run on the new 5.34-petaflops Cheyenne system.

The deadline for NSC submissions has been extended to September 30. NSC allocations target large-scale projects lasting one year to a few years that align with NCAR’s scientific priorities and strategic plans. NSC projects will have access to the Cheyenne system early in 2017.

For more information, see the 2016 Call for Proposals from NCAR Researchers for ASD and NSC Resources.

The GLADE centralized file system returned to service on Wednesday with the home space quota expanded from 10 to 25 GB and block size increased from 512 KB to 1 MB. The larger block size and SSD storage of metadata significantly improved the file system’s performance, most notably for large sequential I/O operations and file metadata requests.

The larger /glade/u/home file space holds dual copies of users’ data for improved data integrity and safety. As a result, the output of users’ gladequota or du (disk usage) commands will show a home space quota of 50 GB and twice the disk space used by single copies. The results of an ls, or other command that deals with file sizes, will continue to report the expected size.

The larger GLADE file system will serve the new Cheyenne supercomputer as well as the Yellowstone system.

NCAR researchers and computational scientists are encouraged to submit requests for Accelerated Scientific Discovery (ASD) and NCAR Strategic Capability (NSC) projects to be run on the new 5.34-petaflops Cheyenne system.

The deadline for ASD submissions is August 29. ASD projects will run on the new Cheyenne system from approximately January through March 2017. The ASD projects will have early access to the Laramie test system in November, early access to Cheyenne in December 2016, and full-scale access in January 2017.

The deadline for NSC submissions has been extended until September 30. NSC allocations target large-scale projects lasting one year to a few years that align with NCAR’s scientific priorities and strategic plans. NSC projects will have access to the Cheyenne system early in 2017.

For more information, see the 2016 Call for Proposals from NCAR Researchers for ASD and NSC Resources.

The CISL Consulting Services Group will present a 90-minute class at 2 p.m. MDT on Tuesday, August 30, to familiarize participants with the Globus system for transferring data between file systems.

Much of the work that scientists do involves large quantities of data that must be transferred for purposes of sharing and collaboration. Globus is a tool designed to facilitate fast, reliable, secure file transfers between supercomputing centers as well as local file systems.

In this class, we will explore using Globus to move files to and from the GLADE file systems. Specifically, we will cover:

  • When to use Globus

  • How to set up a Globus ID and access GLADE

  • Globus web interface vs. command-line interface

  • Creating a Globus endpoint

  • The NCAR Data Sharing Service

There will be time for interested attendees to apply the above information to their workflow.

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

CISL will make dedicated, large-scale resources available on NCAR's new Cheyenne supercomputing system from January through March 2017 as part of its Accelerated Scientific Discovery (ASD) initiative for university and NCAR research. Researchers in the atmospheric and closely related sciences who are supported by the National Science Foundation (NSF) are eligible.

Proposed ASD projects must study a challenging and important problem in the atmospheric sciences or computational science in support of the atmospheric sciences. ASD projects must also:

  • Require 10 million core-hours or more for the set of computational experiments proposed.
  • Utilize production-ready codes that can scale effectively and efficiently up to thousands of processors or efficiently use an ensemble of comparable size.

  • Have the staff time and experience to complete the computational experiments within 12 weeks.

Applications are due August 29, 2016. NSF-supported university researchers who are interested in applying should review these eligibility and proposal requirements for further details. 

CISL is now accepting requests from university-based researchers for large-scale allocation requests; submissions are due September 12. Fall allocation requests should focus on the forthcoming Cheyenne system. Only projects that have previously received Yellowstone allocations and need modest amounts to complete their work should ask for Yellowstone core-hours.

Large allocations for Cheyenne will be those greater than 400,000 core-hours; large allocations on Yellowstone are those for more than 200,000 core-hours. CISL accepts requests from university researchers for large-scale allocations every six months.

See the University allocations web page for more information. The submission instructions include information about estimating Cheyenne core-hour needs.

Please email cislhelp@ucar.edu for questions.