Project

General

Profile

Topics to discuss week of 14-Jan-2013

Things that we should talk about when Alessandro is at Fermilab.

  1. status of group 1 activities
  2. discuss group 2 perpective
  3. discuss PMT initialization
  1. discuss the plan for sorting events and storing them on the online cluster (file systems, etc.)
    • we should sketch out the expected pattern for data filenames on disk (both out of the eventbuilders and the aggregator)
    • [we will meet tomorrow at 09:00 with Jim to flesh this out]
  2. for multi-card readout (e.g. 1495 and TDC), our proposal is to have a special FragmentGenerator derived class that handles this. Is this OK?
    • [One boundary condition is that there should be two separate fragments for the two boards (1495 and TDC) in the final event. Blocking issues will be handled by suitable configuration of the boards and ordering of the getNext() calls to the Driver classes. Alessandro is fine with having a list of FragmentGenerators in any one of several locations in the code.]
  3. work through how to stop and restart runs when using real V1720 boards
    • [Steve will modify ds50driver to demonstrate the problem. Then, Alessandro will use that to investigate the problem.]
  4. learn how to get realtime priorities to work (permissions?)
    • [Alessandro will show us the necessary PC configuration that is needed to get this to work.]
  5. when can we test V1495 readout?
    • [Not for a while, but we will need to modify the V1495Driver class to execute the necessary initialization steps after Boris modified the teststand setup this morning]
  6. when would be a good time to move common/reusable code from ds50daq to artdaq?
    • [any time is fine]]
  7. learn about what has been decided about the database
    • [Postgres has been selected, and a consulting company will provide the DB setup and initial schema.]
  8. can we finish work on the board_reader branch and merge it to the develop branch?
    • [yes, soon]
  9. does the message facility have to use SQL as the tranport, or can it still send messages via XMPP and store them in a DB using a central application?
    • [XMPP with later storage in a DB is fine]]
  10. do we need to talk about file transfers out of the online cluster?
    • [no, this is the responsibility of others]
    • Jonghee is curious about the data rate between LNGS and Fermilab. Are there tests that we can run?
  11. we should go over the list of installation questions from Alessandro with Lynn
    • [DONE - 17-Jan-2013]