[Halld-offline] Offline Software Meeting Minutes, June 26, 2013

[Mark M. Ito]

Folks,

Find the meeting minutes posted at

https://halldweb1.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_June_26,_2013#Minutes

and reproduced below.

   -- Mark
___________________________________________________

GlueX Offline Meeting, June 26, 2013
Minutes

      * CMU: Will Levine, Paul Mattione, Curtis Meyer
      * IU: Matt Shepherd
      * JLab: Mark Ito (chair), David Lawrence, Simon Taylor, Georgios
        Vasileiadis, Elliott Wolin, Beni Zihlmann
      * Northwestern: Sean Dobbs
      * UConn: Alex Barnes[?], Richard Jones

Geant4 Conversion

    Richard is ready to start. Mark will make a branch for Richard to use.
    Although there might be changes to the BCAL code and tracking, we
    agreed it was time to go ahead and start.

Planning for Analysis Workshop/Tutorial

    Paul led the discussion.

    A rather detailed [31]draft agenda has been posted on the wiki.
      * Justin, Matt, and Paul M. have been doing planning over email.
      * Input data sets for each stage that requires one will be prepared
        in advance.
      * Justin generating 5 to 10 million Pythia events and working on
        training the BDT on them.
      * Matt suggested that we have an area of the subversion repository
        for the exercises and talks.
      * Work on the agenda will continue.
      * The [32]registration page is up.
      * Coffee and cookies will be served. There will be an $18
        registration fee.
      * Curtis has already purchased 16 GB USB memory sticks for the
        virtual machines.
      * Participants will be encouraged to install VirtualBox on their
        laptops in advance of the workshop. David also suggested posting a
        link to the VirtualBox download site.
      * The disk image will be available for folks who cannot attend live
        or for future use.
      * We will be using Xubuntu 12.04 LTS.

Top-level build scripts

    Mark plugged his scripts to build the entire GlueX software complex
    from scratch on various flavors of Linux, in including installing
    prerequisite packages from the relevant repositories. See [33]his wiki
    page for details.

BCAL Update

    Will showed recent studies on how charged pions are seen by the BCAL,
    including issues with matching of charged tracks to BCAL clusters. See
    [34]his slides for details.
      * Simon told us that the matching criteria are based on empirical
        distributions of the miss distance between cluster and track
        location at the face of the BCAL.
      * There is a long tail in the time difference distribution, even if
        the matching criteria are tightened, out to 10 ns and beyond. This
        needs to be understood.

Data Challenge 2

    Mark has just started resurrecting his scripts from the first data
    challenge. Not much to report there.

    We discussed what needs to be done before we can start.
      * Further BCAL reconstruction improvements
           + Curtis wondered about solving the timing tail problem that
             Will mentioned and the "curving" of the shower profile seen by
             Andrei Semenov. These problems do not have solutions in hand
             at present. We will have to wait and see if we can expect them
             to come on fast enough to include.
           + Matt suggested that one way to go would be to save the
             individual BCAL hits in the REST format so that we could
             either re-do the reconstruction or use the hits in a future
             algorithm that requires access to individual hits. We
             discussed whether these kind of studies need to be included in
             the data challenge or not.
      * Electromagnetic background
           + Although it is easy to turn this background on, David
             cautioned that using a time window large enough to fill the
             drift time window in the tracking chambers might slow down
             execution by a lot. He has already implemented a system where
             a relatively small number of pure background events can be
             mixed into the events of interest at the hit level. This
             problem will need to be revisited before we start.
      * Noise hits in the BCAL
           + David told us this is easily turned on.
      * Tracking improvements
           + Simon told us that there are a few improvements he is working
             on that we might want to include.
      * Expanding photon energy range
           + This is done with a simple configuration parameter change.
      * Tagging trigger results, both for level 1 and 3.
           + We need to look back and see if this has been done already. If
             not, we should include it.

    Curtis proposed that we set a goal of four weeks from today to make
    decisions on these issues and freeze the code. We thought that that
    would be a good goal.

Online Data Challenge

    David, Elliott, and Sean are working on generating raw data in EVIO
    format from Monte Carlo events using the real translation tables that
    Dave is developing. These tables based on Fernando's spreadsheets.
    These will be used to test the online data pathway, including
    monitoring, RootSpy, and possible transfer to the tape library.

Software Review News

    Nothing new.

    Retrieved from
"https://halldweb1.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_June_26,_2013"

References

   31. https://halldweb1.jlab.org/wiki/index.php/July_2013_Tutorial
   32. 
https://misportal.jlab.org/Ul/conferences/generic_conference/registration.cfm?conference_id=COLLAB-GLUEX-JUL2013
   33. 
https://halldweb1.jlab.org/wiki/index.php/Scripts_for_Installing_GlueX_Software
   34. 
https://halldweb1.jlab.org/wiki/images/1/18/Bcal_offline_mtg_20130626.pdf

-- 
Mark M. Ito, Jefferson Lab, marki at jlab.org, (757)269-5295