GlueX Offline Meeting, February 23, 2011

From GlueXWiki
Jump to: navigation, search

GlueX Offline Software Meeting
Wednesday, February 23, 2011
1:30 pm EST
JLab: CEBAF Center F326/7

Agenda

  1. Announcements
    1. New sim-recon release: sim-recon-2011-02-02: Mark
    2. Computing Model Spreadsheet: Mark
  2. Review of minutes from the last meeting: all
  3. Configuring JANA: David
  4. Calibration Database Update: Dmitry
  5. Action Item Review: all
    1. Resolved issues
    2. Outstanding issues
    • To view these links, login first. Anonymous login is available.
  6. Review of recent repository activity: all

Communication Information

Telephone

To connect by telephone:

  1. dial:
  2. enter access code followed by the # sign: 3421244#

Video Conferencing

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2011-1Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2011-1Q/ . You have to be a member of the "halld" Unix group to do this.

Minutes

Present:

  • IU: Jake Bennett, Matt Shepherd, Kei Moriya
  • JLab: Vardan Gyurjyan, Mark Ito (chair), David Lawrence, Dmitry Romanov, Elton Smith, Sascha Somov, Simon Taylor, Elliott Wolin, Beni Zihlmann

Announcements

  1. Mark reminded us of the lastest tagged release of the simulation/reconstruction software: sim-recon-2011-02-02.
  2. Mark called attention to the Computing Model Spreadsheet that was circulated on the email list. Modifications continue based on feedback from the Collaboration.
  3. Elliott announced that Andrew Blackburn will start work again on the ded event display next week. The next goal: have a simple way to connect a running JANA process to the event display, with two-way communication, for example to try different algorithms tracking on a single selected event within an event stream.
    • There was a discussion of how we decide on event display technology and whether we had enough manpower to pursue more than one avenue and, related to that, why is the current viewer (hdview2) not good enough. Andrew has been using the bCNU framework developed by Dave Heddle of CNU. Dmitry has been planning to look at using the EVE package of ROOT, but has not had much time to look at it. hdview2 also uses a subset of ROOT. Dave told us that hdview2 was not designed as a platform for long-term development. Elliott told us that the effort from Andrew and Dave H. is, if not free, of very low cost to the Collaboration and the framework will in any case be used in Hall B. EVE can be easily tied to the HDDS geometry specification (a ROOT version of the HDDS geometry already "exists") and would have built-in 3D capabilities as well as 2D projections. Elton in particular questioned whether we can sustain multiple efforts, the EVE route being the most speculative at this point. We also discussed the question of who would maintain the program. Andrew is an undergraduate and Dave H.'s primary commitment is to CLAS12. MEPhI (Dmitry's institution) might be able to support a software effort, but they are already considering several hardware contributions. David thought that we should convene a meeting sooner than the next offline meeting to discuss the issues, especially since Andrew is slated to start next week. Clearly more discussion is needed; we did not come to any decisions.

Review of minutes from the last meeting

  • The new release of JANA, 0.6.3, has arrived. Note that the build steps have changed slightly from previous versions.
  • David is doing work to characterize the "split-off problem". He intends to feed back what he finds to the folks working on BCAL reconstruction, among others.
  • Beni reported that Richard Jones has checked in some changes to HDDM to eliminate compiler warnings when building with gcc 4.4.5.
  • There was a meeting last week to discuss the way timing is handled in the reconstruction. One take-away that Matt mentioned was the given the way times are generated in the Monte Carlo, where t=0 is defined as the time where the photon beam crosses the center of the target, using the raw times is equivalent to perfect resolution and a perfect identification of the beam RF bucket. We can therefore just go with the time offset in the raw times as if we are in the era of identifying the RF bucket since the RF-bucket timing should be very good compared to any measured time in the detector.
  • Kei will present an update on his PID studies at the next Physics Working Group meeting.

Configuring JANA

David described the system to set parameters, mainly for reconstruction, in the JANA framework. The principal features are:

  • Ability to change the parameters at run-time as an option on the command line
  • Provision for defaults
  • Provision for discovery of parameter settings
  • A system for specifying non-default values of parameters via a configuration file
  • A system for generating comprehensive configuration files from a given instance of a program.

See his slides for details. Kei remarked that the talk had a lot of valuable information and it should be linked from the wiki in an appropriate location.

Calibration Database Update

Dmitry gave a report on recent progress. He described the format of a request string used to access constants from the CCDB. It specifies type of constants, run, variation, and modification time. All can be specified, but only type is required. If omitted, the other fields have default values. He also gave us status summary for the various aspects of the problem. See his slides for details. Dmitry returns to Moscow this weekend, but will be in touch electronically and plans to continue his work on this project.

Action Item Review

We looked at them. See the links in the agenda above.