Daily Bulletin Archive

December 27, 2018

No downtime: Cheyenne, GLADE, Geyser_Caldera, HPSS and Casper

 

December 21, 2018

The Fortran language has evolved a lot since its first version in 1957 in order to take advantage of new computing systems. Now, the use of vector instructions (or SIMD) can result in substantial performance improvements on current and future architectures. Alessandro Fanfarillo, NCAR Research Applications Laboratory, will lead a workshop January 29 to help participants make their Fortran codes run more efficiently through vectorization and other techniques.

When: 9 a.m. to noon, 1 to 3 p.m. Tuesday, January 29

Where: Room 3131, Center Green campus (CG1), Boulder

Participants will get a detailed, practical explanation of how to obtain high performance from modern Fortran codes, with a particular focus on how to exploit the hardware instructions provided by modern processors. Prerequisite: Basic knowledge of Fortran 90 constructs, such as array syntax and allocation, recursion, modules, and intrinsic, elemental and pure functions. Participants are encouraged to bring their own codes and laptop computers.

Register here to attend.

December 21, 2018

Some CMIP6 data are now available on GLADE through the CMIP Analysis Platform, and users can request the addition of other data sets as modeling centers publish them. NCAR's initial CMIP6 data products are expected to be available in the first quarter of calendar year 2019.

More information: CMIP Analysis Platform.

December 20, 2018
A brief outage of the Slurm scheduler will occur at noon today to improve the prioritization of job resource requests on the data analysis and visualization nodes. The work should last up to one hour. During this time, Slurm scheduler commands may be unavailable and new jobs may not be submitted. Pending and running jobs should be unaffected. Users will be notified when the work is complete.
 
 
December 20, 2018

The 2019 Improving Scientific Software conference organized by the UCAR Software Engineering Assembly (SEA) will take place April 8 to 12 in Boulder, Colorado. The SEA is soliciting talks and tutorials about any aspects of scientific software, but particularly on the following topics:

  • How to improve scientific software

  • Utilizing modern HPC architectures for scientific software

  • Utilizing machine learning techniques in scientific software

  • Utilizing software engineering practices to improve scientific software

  • Utilizing containers and cloud for HPC

The deadline for submitting abstracts is January 13. Students can apply for travel assistance until February 9. See the conference site for more details.

December 19, 2018

The Geyser and Caldera clusters will be decommissioned on December 31, as announced previously. Users who have not already done so should migrate their data analysis and visualization (DAV) workflows to the new Casper cluster. See the following links for information:

December 17, 2018

No downtime: Cheyenne, GLADE, Geyser_Caldera, HPSS and Casper

December 12, 2018

GLADE users who have not already done so need to take appropriate action soon to copy any files they need from the /glade/p_old/ or /glade/p_old/work file spaces. As announced previously, they will be decommissioned December 31 and all files will be removed permanently.

Users who have files in those spaces that they need to retain should copy them to one of the new storage systems. See Transfers between GLADE file spaces for how best to do this.

Project spaces

CISL recommends moving active project data to /glade/p/<entity>/<project_code>where entity can be univ, uwyo, cesm, mmm, nsc, or other designated NCAR lab or special program. A one-year purge policy will be enforced on files in those new spaces, meaning files that are not accessed for more than one year will be deleted.

Project data that are not active but need to be preserved should be moved to the Campaign Storage archive. Users access and manage their Campaign Storage files with Globus services. A five-year purge policy will be enforced on Campaign Storage effective from the date files are created in that archive.

Work spaces

All users now have individual directories in /glade/work with 1-TB quotas. Files in those directories are not purged. Users should copy any files they need from their /glade/p_old/work/ directories to their new /glade/work directories before December 31.

December 11, 2018

CISL staff will update several key data analysis and visualization system components and the Campaign Storage file system today. The maintenance will affect the availability of the Casper, Geyser, and Caldera clusters and file transfers to Campaign Storage. Cheyenne and the GLADE file system will remain available throughout the day.

Maintenance on the Campaign Storage file system will begin at 9 a.m. MST and is expected to last until 5 p.m. MST. Globus transfers will be suspended for the duration of the maintenance but will resume when the work is completed.

Maintenance on Casper, Geyser, and Caldera will begin at 9:30 a.m. MST and is expected to take approximately three hours. The work will include installing a new version of the Slurm resource scheduler. These clusters will not be accessible during the maintenance window. Also note:

  • All running jobs, interactive processes, and login sessions will be terminated when maintenance begins.

  • Jobs that are queued for execution or in a hold state will remain in those states until Casper, Geyser, and Caldera are returned to service.

December 11, 2018

CISL system administrators were able to install a new version of PBS during an unscheduled Cheyenne outage on Tuesday, December 4, so no Cheyenne downtime is needed during the maintenance period scheduled for December 11. Other system maintenance remains scheduled for December 11 as follows.

CISL staff will update several key data analysis and visualization system components and the Campaign Storage file system beginning at 9:30 a.m. MST. The maintenance will affect the availability of the Casper, Geyser, and Caldera clusters and file transfers to Campaign Storage. The GLADE file system is expected to remain available throughout the day.

Maintenance on Casper, Geyser, and Caldera is expected to take until 12 noon MST. The work will include installing a new version of the SLURM resource scheduler. These clusters will not be accessible during the maintenance window. Also note:

  • All running jobs, interactive processes, and login sessions will be terminated when maintenance begins.

  • Jobs that are queued for execution or in a hold state will remain in those states until Casper, Geyser, and Caldera are returned to service.

Maintenance on the Campaign Storage file system is expected to take until 5 p.m. MST. Globus transfers will be suspended for the duration of the maintenance but will resume when the work is completed.

Pages