Shared computing platforms

We use two primary shared computing platforms:

  • Stanford Research Computing Facility

  • Texas Advanced Computing Center

The decision regarding which platform to use will depend upon the specific needs of the project, and can be discussed with Dr. Poldrack or other experienced researchers in the lab.

Shared computing etiquette

Lab members should follow good shared computing etiquette:

  • Never run a computing job on the login node

    • For testing purposes, you can access a development node.

  • Keep in mind that our allocation on these systems is limited.

  • Run a small test job before starting a large job

  • Ensure that your jobs are efficiently using the system resources - if you aren’t sure, then ask for help!

  • On systems where the lab has a specific allocation of cores (e.g. Sherlock), don’t use all of the cores at once unless having first discussed your needs at the lab meeting or via the lab mailing list or Slack.

  • Be sure to set up your environment properly:

    • Always set your umask to allow group read/write; otherwise you may create files that can’t be modified by others in the group.

  • Share your wisdom

    • If you encounter a problem that is not documented in the lab’s internal documentation, then add it to the documentation once you have solved it

    • If you encounter an error in the documentation, then flag it and/or fix it

    • You can refer to this website for more information on Sherlock, and if you have any further questions you can contact the Sherlock admis via email/Slack or meet during specific Office Hours as specifiedhere.

  • Each TACC system has its own dedicated user guide that provides substantial information about how to use the system.

  • Users of shared systems should clean up directories as projects are completed, removing intermediate files that are not necessary to keep. The storage of redundant data should be avoided, except where required for backups.