[Halld-offline] Offline Software Meeting Minutes, August 7, 2013

[Mark M. Ito]

Folks,

Find the minutes below and at 
https://halldweb1.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_August_7,_2013#Minutes 
.

   -- Mark
_____________________

GlueX Offline Meeting, August 7, 2013
Minutes

    Present:
      * CMU: Paul Mattione, Curtis Meyer
      * JLab: Mark Dalton, Mark Ito (chair), David Lawrence, Simon Taylor,
        Beni Zihlmann
      * MIT: Justin Stevens
      * UConn: Alex Barnes

Announcements

      * Mark reminded us about the [26]new version of CCDB that we are
        using now: 0.08. He also flashed the [27]new wiki page that will be
        used to keep track of CCDB versions.
      * Mark also pointed out thet there is a tagged version of sim-recon
        that corresponds to the version built on the virutal machines at
        the recent Analysis Workshop. The tag is
        tags/sim-recon_analysis_ws_2013 . Versions of the other packages
        are documented on the [28]Workshop page.

Review of Minutes from the Last Meeting

    We looked over the [29]minutes of the July 24th meeting.
      * Paul has collected a [30]list of analysis channels from the
        collaboration. This will be discussed at Monday's Physics Meeting.
      * We will postpone the second data challenge (DC2) and let the online
        data challenge go ahead. It is scheduled to begin at the end of
        this month. Mini-challenge jobs should continue to run however in
        preparation for full production.
      * If Geant4 is ready in time for DC2 we may use it. We would probably
        not delay the start for it, however.
      * Since the EM background is going to be added to the b1pi bi-weekly
        jobs we can use those tests to judge how to include EM background
        in DC2.
      * We should change the scripts to run the b1pi tests in parallel on
        the different platforms. Same goes for the single-track tests.
      * The date problem in the b1pi scripts has had one iteration of fix
        attempt by Mark.

Very Raw Data Classes

    David is starting to institute a layer of low-level objects below the
    "hit" objects that we have been using in the simulated data when
    analyzing the very raw data in EVIO format from the electronics. These
    objects are distinguished by having "digi" in their names. Factories
    will be written to take the digi objects and transform them to our
    traditional "hit" objects. If the "hit" objects are present in the
    input data file, no digi objects nor their associated factories
    invoked.

    David will explain his scheme further at the next Offline Meeting.

Work Tools and Data Provenance

    JLab Scientific Computing has organized a meeting to discuss what they
    can do to help the ENP effort in two areas:
     1. Work Flow Tools: standard automation of large-scale, multiple job
        submission and tracking
     2. Data Provenance: having information on how data was processed and
        archiving tools necessary to re-analyze data in the future.

    The first meeting is tomorrow at 3 pm. David and Mark will represent
    GlueX.

    Retrieved from
"https://halldweb1.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_August_7,_2013"

References

   26. 
https://mailman.jlab.org/pipermail/halld-offline/2013-July/001352.html
   27. https://halldweb1.jlab.org/wiki/index.php/Tagged_Releases_of_CCDB
   28. 
https://halldweb1.jlab.org/wiki/index.php/GlueX_Analysis_Workshop_2013#Software_.2832-bit_builds.29
   29. 
https://halldweb1.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_July_24,_2013#Minutes
   30. https://halldweb1.jlab.org/wiki/index.php/Physics_Analysis#Analyses
______

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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20130807/1be0821f/attachment-0002.html>