Status as of Monday, 4/21/15

Meeting with Jon and Ram, with Oliver and Michael.

Overall

Group members verified value in keeping both workstations and retaining their functionality. Invest to upgrade OS and data storage so more robust and maintainable indefinitely.

  • Current server has a failed hard drive.
  • Current OSes are outside vendor support.
  • Current set-up does not promote good practices regarding data stewardship.

Decided

  • Get $9/mo SFS (120GB) NFS.
    • Jon to get Oliver an account number this week. This expense is eligible for using restricted funds.
    • ChemIT will do the paperwork and technical configuration, timed to when Lulu can start using the service.
    • No data will be migrated there by ChemIT.
      • Group expects that members will add and retain only the data they are actively working on.
  • Lulu to take one of the workstations (notifying Ram, and Jon) to work on (within a week or two?). Lulu will:
    • Put new OS (CentOS 6 or 7) and ensure 3-D visualization software an be made to work with monitors and active 3D glasses on this new OS. If not, new plan. Otherwise, continue.
    • Identify any non-SBGrid software required.
      • For each required application, identify the group member responsible for verifying function of software during later testing phase of project.
      • Lulu to verify required software runs on selected upgraded OS.
    • Confirm best method for logging in. Likely by Cornell NetID/ password credentials.
    • Connect the workstation to SFS's NFS share and ensure home directories reside there.
    • Set up station for group members to test and verify software and function.
  • Group will test new workstation.
    • Lulu will ensure old data is available to researchers.
      • Likely via group-provided USB-connected device able, which researchers can elect to connect to either workstation, one at a time.
    • Timing: Can this be completed within a week.

If all is confirmed to be working as needed, Lulu will upgrade the second workstation. And we'll re-purpose or dispose of the failing server, per group's preference.

  • No labels