Daily Bulletin Archive

January 15, 2020

Users are reminded that the High Performance Storage System (HPSS) tape archive will become read-only on January 20. The only exceptions will be for curated collections such as the Research Data Archive. This follows the October 2019 announcement that HPSS will reach its end of life and be decommissioned in 2021.

While writing HPSS files will not be possible as of January 20, users will still be able to perform most common meta-data operations on their HPSS holdings, including deleting, renaming, and moving files. Those who have not already done so should begin deleting files that are no longer needed and moving other data to alternative storage systems.

CISL recently published documentation and training for users on recommended processes for identifying and organizing HPSS holdings; copying files that need to be preserved to another storage resource; and deleting files that are no longer needed. Follow these links:

Additional CISL Daily Bulletin announcements will follow as new tools, training, and expanded documentation become available to support the data migration. In the meantime, please contact CISL for advice on individual workflows and storage options.

January 13, 2020

No scheduled downtime: Cheyenne, Casper, Campaign Storage, GLADE, HPSS

January 13, 2020

The CISL User Services Section will present an in-person and online tutorial at 10 a.m. MST on Thursday, January 23, for general users who want to know more about the Casper data analysis and visualization cluster. There are no prerequisites.

A 30-minute presentation by Shiquan Su of the CISL Consulting Services Group will provide attendees with a basic understanding of how to run jobs on the Casper cluster. The presentation will be followed by additional time for answering questions. Topics to be covered include:

  • Node types for different tasks

  • Accessing Casper resources through the Slurm job scheduler

  • Interactive jobs and virtual remote desktops

  • Using GPUs on Casper

Register to attend in person at NCAR’s Foothills Lab in Boulder (FL2-1001) or attend online by selecting one of these links:

January 13, 2020

The Cheyenne cluster will be unavailable to users in late February to allow CISL staff and HPE engineers to perform critical infrastructure maintenance, repairs, and system software updates. The maintenance period will begin Tuesday, February 25, and the system is expected to be returned to service no later than Saturday, February 29.

The Casper cluster, GLADE file systems, Campaign Storage, data-access nodes, and HPSS will remain available throughout the week. Watch for more details and updates in upcoming Daily Bulletins.

January 10, 2020

The Artificial Intelligence for Earth System Science (AI4ESS) Summer School is accepting applications until January 31. The event will take place the week of June 22-26, 2020, at NCAR's Mesa Laboratory in Boulder.

Who should apply

Graduate students, postdocs, faculty, research scientists, software engineers, and data scientists focused on Earth system science are encouraged to apply. Previous machine learning experience is not required. Some experience with Python and scientific Python libraries will be needed to contribute to the breakout sessions.

Application process

We have room for up to 80 participants but are expecting interest to exceed capacity. In order to ensure that the applicants who could most benefit can attend, we have a formal application and review process for the summer school. To apply, please fill out the application form. Applications are due by Friday, January 31, at 5 p.m. MST. Acceptance notifications will be sent out in early March, and those selected will be invited to register. A waiting list will also be created in the event that some participants are not able to attend.

January 6, 2020

No scheduled downtime: Cheyenne, Campaign Storage, GLADE.

Scheduled downtime for Casper login nodes from Jan 7th 12 p.m. - Jan 7th 2 p.m.

Users are advised to use the Cheyenne login nodes to access Casper during this time.

January 2, 2020

Individuals who use the ExtraView help ticketing system are now required to log in with a CIT (Active Directory) password rather than their UCAS (Kerberos) password. The change, which was announced recently in Notifier emails to users, took effect this morning.

Most CISL help desk support is now provided via the Staff Support and NCAR Research Computing support portals rather than ExtraView, which NETS, Infrastructure, and Asset Managers still use.

December 31, 2019

The Casper node upgrades described in this previous article have all been completed. New hardware required updates to network software on all Casper nodes. The updates required Open MPI 3.1.4 to be rebuilt and it is likely that older versions of OpenMPI will no longer work. CISL will test existing Open MPI modules in the coming weeks and remove any versions that no longer function correctly.

December 31, 2019

Registration is now open for an NCAR/CISL series of five one-day Fortran workshops beginning Tuesday, January 7. Dan Nagle, CISL Consulting Services Group software engineer and a member of the U.S. Fortran Standards Technical Committee, will provide the training at the Center Green Campus (CG1-3131) in Boulder.

Participants are encouraged to bring their own laptop computers with recent releases of gfortran, mpich, and opencoarrays. Each workshop will begin at 9 a.m. and end at 4 p.m. with an hour break at noon.

  • Scalar Fortran - Tuesday, Jan. 7: Scope and definition of names, how a Fortran program interacts with the o/s, intrinsic types, inquiry procedures.

  • Vector Fortran - Tuesday, Jan. 14: Arrays and array operations, rank of an array, inquiry procedures, array intrinsics.

  • Object-Oriented Fortran - Tuesday, Jan. 21: Derived types, operators, assignments, polymorphism.

  • Parallel Fortran - Tuesday, Jan. 28: Coarrays, solutions to several problems explained in detail.

  • Interoperability with C and handling Floating Point Data - Tuesday, Feb. 4: How to write and call libraries written in C  from Fortran; IEEE 754 floating point: rounding modes and exceptions.

Use this form to register to attend one or more workshops. The workshops will not be webcast or recorded.

December 30, 2019

Several Casper nodes were updated with new high-speed network hardware earlier this week. The new hardware required updates to the system’s network software on those nodes which includes all nodes with NVIDIA V100 GPUs.  The majority of Casper’s nodes have not yet been updated and their network is unchanged. The list of updated nodes is provided below. 

The updates required Open MPI 3.1.4 to be rebuilt and it is likely that older versions of OpenMPI will no longer work on the updated nodes.  Intel MPI usage should be unaffected by the network upgrade. However, until the remainder of Casper’s nodes are updated with the new network hardware, multi-node jobs with a mix of updated and non-updated nodes will likely perform slower than expected using any MPI library.  Further changes to Casper’s OpenMPI software stack are likely in the next several weeks. Please watch for related announcements in the Daily Bulletin.

Updated Casper nodes: casper08, casper09, casper23, casper24, casper25, casper27, casper28