GlueX Offline Meeting, August 8, 2012

From GlueXWiki
Jump to: navigation, search

GlueX Offline Software Meeting
Wednesday, August 8, 2012
1:30 pm EDT
JLab: CEBAF Center, F326/327

Agenda

  1. Announcements
    1. Welcome back, Dmitry!
    2. New release: sim-recon-2012-07-26
    3. Scientific Computing Survey
  2. Review of minutes from the last meeting: all
  3. Report from Data Challenge Meeting
  4. Reconstruction sub-group reports
    1. Calorimeters
    2. Tracking
    3. PID
  5. Changes to bggen to support REST rationality
  6. Action Item Review
  7. Review of recent repository activity: all

Communication Information

Video Conferencing

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2012-3Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2012-3Q/ .

Minutes

Present:

  • CMU: Will Levine, Paul Mattione, Curtis Meyer
  • IU: Ryan Mitchell
  • JLab: Mark Ito (chair), David Lawrence, Elliott Wolin

Announcements

  • Dmitry Romanov is back at JLab. He will be here for the next six months.
  • Mark put out a new release: sim-recon-2012-07-26.
  • We will look into upgrading some of our support packages including Xerces-c, ROOT, and Jana.
  • Curtis reminded us of the recent build errors. Mark will investigate.
  • Mark reminded us to fill out the Scientific Computing Survey!
  • Elliott announced that next week we will upgrade our MySQL server. This will cause a short interruption of access to the wiki. Related to this we discussed having multiple servers to avoid outages of critical applications, and what their purposes should be. Discussion will continue.
  • Mark announced that, in fact, we do have a new MySQL server, hallddb1.jlab.org. It will sit outside the firewall, network-wise. It will be used to serve out the CCDB and will be available read-only both inside and outside JLab.
  • Elliott told us the Online Group is working on the raw data tape. David Abbott has provided some documentation on the API-to-come to support the writing of raw fake data. The initial version will produce data with one event per block (no entanglement).
  • Related to this, David has started work on code to do the EVIO to DANA conversion for the low-level raw data.

Review of minutes from last meeting

We reviewed the minutes from the July 25 meeting.

David continues to work on the detailed digitization scheme. He has committed a new data model, with hooks for this effort. He found an issue with the energy dependence of the energy resolution which he traced to a threshold problem with the ADC data. He fixed this by allowing different thresholds for the TDC data and ADC data. The data model needs to accommodate two time measurements as well, one from each.

Report from Data Challenge Meeting

Mark reviewed the minutes from the July 30 meeting for us.

Kinematic Fitting

Paul has finished writing the new kinematic fitter. It does 4-momentum conservation with results identical to the old fitter. It has not been tested on long decay chains and timing has yet to be incorporated.

REST

Paul has been using the REST format to do analysis for strangeness in the context of the PID upgrade. The results are not identical to those from analysis without using the format, but close. For example he sees 2% more events passing PID cuts with REST. He has been writing a bunch of code for analysis; he hopes to have an update ready for the next Physics Meeting.

New Action Items

  1. Upgrading some of our support packages including Xerces-c, ROOT, and Jana. -> David, Mark
  2. Discuss multiple database servers, and what their purposes should be.