Agenda:

 

  1. Transitioning Jeremy’s tasks.                           — Jeremy, All
    1. Software Releases.
      1. Make a new software release this week with someone else.
      2. Document.
    2. Conditions System
      1. Is the current documentation adequate?
      2. New managers:
        1. Who will do SVT calibrations?
        2. Who will do the ECAL?
      3. New maintainer:
        1. Extend system for Hodoscope.
    3. Monitoring system maintainer.
      1. Updates and additions for next run.
    4. HPS-MC scripts.
      1. JLab production should switch to these scripts.
      2. Verify compatibility.
    5. Data Catalog
      1. Improvements needed?
      2. Folder structure.
      3. Additional meta data.
      4. Incorporate in MC scripts.
    6. HPS-SIM code.
      1. Development of SLIC replacement.
    7. lcsim fixes.
      1. Who has access?
      2. Who has this knowledge?
  2. Questions/issues?                                                            -- All

 

Notes:

  • Jeremy will leave starting April 2:
    • April 2 - May    – "Vacation" - still access to accounts.
    • May - Octobe   – Casual employment - will still have access, and do some tasks, can still help.
  • Transition Plan on Confluence.
  • HPS-Java and LCSIM
    • LCsim goes through the SLAC lab GitHub group. Needs a SLAC account.
    • We should not need to make major modifications. If really needed, there are people at SLAC who can help.
  • Software Releases.
    • There is a document on this in the "deprecated" section of Confluence. Needs a cleanup.
    • Make sure the tests work before making a release.
    • Occasionally a (scary) "git rebase" is needed. 
    • Jeremy will make a new release together with Norman.
  • Conditions System:
    • There is administration: 
      • Coordinating updates of the DB.
      • Managing access.
      • Dealing with emergencies (i.e. revert to a backup.)
    • Loading calibration data: This does not have good documentation.
      • We will need to do this for the next run!!!
      • Command line tools to load the data.
      • Need more on the codes that create the calibrations.
      • Database and API will need to be updated for the hodoscope.
  • Detector Simulation:
    • SLIC:
      • Mostly stable. Probably better not to touch this code. Consider this frozen.
    • hps-mc:
      • Used at SLAC.
      • JLAB should switch to this tool. Avoid duplication. Bradley will work on switching to this tool.
      • Needs documentation.
    • hps-sim:
      • This would be the future of SLIC. We should put new development here.
      • Needs verification against SLIC output.
  • Run Database:
    • We need to decide what to do with this. 
    • Do we want to deprecate this, replace this, carry forward.
    • In future meeting – We should discuss the use cases, see how this needs to change to be useful.
  • Monitoring Application:
    • Updating and testing (fixing) before physics run.
    • Spaghetti code.... careful if you want to change this.
    • Need to add the hodoscope to the app. Should be similar to ECAL. Can Kyle do this?
    • Low level stuff: you would probably be better off writing a separate application.
  • Data Catalog:
    • We have a lot of data on disk. (3.5 million MC files!)
    • Core tools are externally maintained.
      • Python scripts are HPS specific to load the data.
      • Data load should happen from the MC scripts and recon scripts.   Bradley looks into this.
    • We need to come with a better naming structure for the MC output. 
      • Tape archive is static, so we cannot rename files already on tape. Bradley looks into this.
  • Next Run:
    • We really will need to be on the ball!
  • Questions/Concerns:
    • How difficult is it to change/modify the database?
    • Are there incremental/full backups made of this database by JLab?
      • This needs to be confirmed.

HPS Transition Plan.pptx



  • No labels