[Halld-offline] Offfline Software Meeting Minutes, June 25, 2014

[Mark M. Ito]

Folks,

Please find the minutes below and at 
https://halldweb1.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_June_25,_2014#Minutes 
.

   -- Mark
_______________________________


GlueX Offline Meeting, June 25, 2014
Minutes

    Present:
      * CMU: Will Levine, Paul Mattione
      * JLab: Mark Ito (chair), David Lawrence, Simon Taylor, Beni Zihlmann
      * IU: Kei Moriya
      * NU: Sean Dobbs
      * UConn: Alex Barnes[?], Richard Jones

Announcements

      * Mark mentioned that CCDB 1.02 will be released soon. Sean explained
        that it fixes a bug that appears when using a non-default
        variation.
      * Mark also mentioned that we need a new tagged release soon.
      * We discussed whether EVIO should be made a mandatory part of the
        build. Most people that that that would only add unnecessary
        complication at this point. We should be prepared to include it in
        builds at JLab in the near future.

Review of Minutes from the Last Meeting

    We skimmed the [19]minutes from June 21.
      * Mark will start copying data challenge 2 data from Northwestern
        when the servers come back up.
      * David told us that as soon and the TOF group can obtain data using
        the CAEN TDCs he can test that aspect of reading EVIO data into
        JANA. Beni mentioned that he has already sent David some F1-TDC
        data and so that can be tested as well.

BCAL Timing

    Will gave us an update. He showed a [20]couple of plots showing a
    comparison of the late-time tail for positive vs. negative pions and
    the timing tail for protons. These studies were at least partially
    motivated by [21]Richard's hypothesis that the tail could be a result
    of metastable states produced from stopped pions.

    We spent some time re-studying [22]the hdgeant event he sent to the
    email list some weeks ago. Will will package a few similar events and
    send them to Richard so he can look more carefully at the physics
    treatment of the stopped pion.

Data Challenge 3

    We discussed possible goals for the data challenge:
      * process fake raw data resident in the tape library
      * render the date in EVIO format
      * perform more efficient generation of the electromagnetic background
      * use the Geant4 version of HDGeant
      * incorporate new code for obtaining constants from the CCDB

    Richard thought that the mixing of E&M background should happen at the
    simulation stage. He outlined how he thought that should go: pick out
    the appropriate binary tree from a background event and insert it into
    the event of interest, then let the normal hit-making mechanism
    proceed. We would need to find someone to work on this; Richard does
    not have time right now.

    Paul pointed out that if all we interested in is testing tape access,
    then many of the other goals could be deferred.

    We discussed schedule. If we schedule it after the summer, the time is
    already short and David pointed out that there will be a
    manpower-availability conflict with the commissioning run. If we wait
    until early next year, we may not learn much beyond what real raw data
    will teach us.

    With all that in mind, we thought that a data challenge with scaled
    down goals for mid-August sounded like a reasonable approach.

    [Added in press:] Curtis thinks that the tape-resident data and EVIO
    format should be included. He also pointed out that this will be a
    chance to exercise JANA multi-threading. He agreed that a mid-August
    timeframe was reasonable.

Action Items

     1. Make a new tagged release. -> Mark
     2. Add EVIO to JLab builds as an option -> Mark
     3. Assign EM background generation problem to someone. -> Mark

References

   20. https://halldweb1.jlab.org/wiki/index.php/20140625_BCAL_Timing
   21. 
https://mailman.jlab.org/pipermail/halld-offline/2014-June/001686.html
   22. https://mailman.jlab.org/pipermail/halld-offline/2014-May/001681.html