Page tree
Skip to end of metadata
Go to start of metadata

Welcome the Smithsonian Institution High Performance Computing Wiki.

This Wiki holds information for the use of HPC resources at the Smithsonian.

Central to the SI HPC resources is the Smithsonian Institution High Performance Cluster (SI/HPC), named Hydra.

  • High performance computing is administered by the Office of Research Computing.
  • The OCIO data center in Herndon, VA houses the high performance computing cluster, Hydra.

What's New

  • Dec 4, 2018

    • We have disabled the cluster's oldest compute nodes (i.e., compute-6-*) from the queues and will retire these 8-year-old machines (12 machines, 16 CPUs, 32GB of memory each).
    • We have acquired 16 new machines (40 CPUs, 384GB each). These new machines need an OS upgrade (from CentOS 6.9 to 7.x) to come on-line.
    • We are in the process of configuring and validating our next OS upgrade. We hope to implement it during the 1st quarter of 2019. We are migrating to a different configuration manager, but will make every effort possible for the cluster to look as much as possible the same after that upgrade and migration (CentOS 6.9 to 7.x, Rock 6.2 to BCM 8.x).
    • We are also evaluating new storage systems to increase the available disk space on Hydra, and continue looking for funding sources to pay for it. Since these evaluations are taking more time than expected, we are considering adding some disk space to our current system (NetApp). If you are interested in acquiring project specific disk space, and have funds you can allocate for that purchase, please contact Sylvain (at hpc@cfa.harvard.edu) by December 10th 2018 to get more details.
  • Nov 5,2018

    • We are very pleased to announce that the systems administrator duties for Hydra will be supported by Dajiang (DJ) Ding effective immediately. DJ supported the cluster half time for many years, but has recently become available to support us full-time. He will be taking over for Jamal Uddin, who has left the Institution. From now on, DJ will be responsible for administering and maintaining the HPC resources at HDC (i.e. Hydra, full time). He will also participate in enhancing Smithsonian’s high-performance computing infrastructure and services. 

    • We ask that you submit HPC sys-admin related issues (forgotten password, something is not working anymore, etc) to SI-HPC-Admin@si.edu. Bioinformatics/Genomics specific questions/requests (request to install or how to run specific s/w) should be sent to SI-HPC@si.edu, SAO users who need help for their applications should contact Sylvain at hpc@cfa.harvard.edu. This will help anyone from the SI/HPC support team to address your issues as soon as possible. Please let us know if you have any questions.

  • Reminders
    • The NVIDIA OpenGL libraries have been loaded on the login nodes.
    • The default Intel and PGI compilers have been changed to use the most recent ones.
    • New versions of compilers and IDL are available on Hydra.
    • An interactive queue has been added to Hydra, interactive use of the login nodes is being monitored.

    • All public disks (/pool and /scratch only) are scrubbed. Reasonable requests to restore scrubbed files must be sent no later than than Friday following the scrubbing, by 5pm.

    • Two test queues are available:

      • uTSSD.tq - allows use of a local SSD for jobs performing heavy I/Os,

      • uTGPU.tq - allows use of a GPU for code built with GPU capability.

  • Past news with more details can be found at the What's New page.

 FYI

  • References to Hydra (publications, proposals, etc) should mention the cluster as:
the Smithsonian Institution High Performance Cluster (SI/HPC).

Last updated  / PBF/SGK/MPK.

Search this documentation


Recently Updated Pages

  • No labels