GlueX Offline Meeting, September 18, 2013
GlueX Offline Software Meeting
Wednesday, September 18, 2013
1:30 pm EDT
JLab: CEBAF Center F326/327
- 1 Agenda
- 2 Communication Information
- 3 Minutes
- Software Review: November 25 and 26
- Data Management Plan
- Review of minutes from the last meeting: all
- Report from Work Flow meeting (Mark, David)
- October Collaboration Meeting
- Software Review Planning
- performance tracking
- summary histograms from DC 1
- Update on Generated Information (Kei) - last week's Particle Decay Chain Meeting
- Data Challenge 2 (Mark)
- b1pi and single track test failures (Mark)
- Mantis Bug Tracker Review
- Review of recent repository activity: all
- ESNet: 8542553
You can view the computer desktop in the meeting room at JLab via the web.
- Go to http://esnet.readytalk.com
- In the "join a meeting" box enter the Hall D code: 1833622
- Fill in the participant registration form.
To connect by telephone:
- US and Canada: (866)740-1260 (toll free)
- International: (303)248-0285 (toll call) or look up toll-free number at http://www.readytalk.com/intl
- enter access code followed by the # sign: 1833622#
Talks can be deposited in the directory
/group/halld/www/halldweb1/html/talks/2013-3Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2013-3Q/ .
- CMU: Paul Mattione, Curtis Meyer
- IU: Kei Moriya, Matt Shepherd
- JLab: Mark Dalton, David Lawrence, Mark Ito (chair), Nathan Sparks, Simon Taylor, Beni Zihlmann
- MIT: Justin Stevens
- UConn: Alex Barnes
- The second 12 GeV Software Review will be held November 25 and 26 at JLab.
- We have been asked by Graham Heyes to prepare a Hall-specific data management plan. Mark has sent him a draft based heavily on the template that Graham has provided.
Review of minutes from the Last Meeting
We looked at the minutes from the September 4th meeting. Simon reported that he has started using the new code from Paul that does a better job of handling curling tracks in the CDC.
Report from Work Flow Meeting
David and Mark attended another work flow meeting, held September 17. More preliminary discussion about Hall practices; not much to report.
October Collaboration Meeting
We went over the proposed agenda for the offline software sessions.
- Will Levine will give the talk on BCAL reconstruction progress.
- Paul will cover both the report from the Analysis Workshop and the standard ROOT trees generated from his analysis package.
- Justin will re-cap the use of his BDT software and may cover topics from a recent paper from him and Mike Williams on training the BDTs to be not only efficient but flat in phase space.
- Times will have to be re-adjusted.
Software Review Planning
Curtis led us through the wiki page he and David started to collect information on the upcoming review. Items from the discussion:
- Overall Curtis thought that we were in pretty good shape given the progress we have made since the last review.
- Matt reminded us that we were charged with presenting the user's point of view at the last review. Ideas were floated on how to do that: video testimonials, experience at the Analysis Workshop, the notion that the software is so easy to use that even Jo Dudek can do a partial wave analysis.
- Rolf Ent had convened Division-wide meetings in advance of the last review. We have not heard of plans for such meetings this time.
- Curtis may pump Rolf for more information on what we should expect.
- Last time we had weekly phone conferences for review planning. We ill fire those up after the collaboration meeting.
- Mark pointed out that we have not done the periodic code profiling the committee asked for. Some simple form of this should be do-able in advance of the review.
- Justin volunteered to look at making plots from the data challenge data to visually display the statistics we generated. Matt suggested looking at, say, an ω peak before and after BDT criteria are applied.
Update on Generated Information
Kei followed up on his presentations at two meetings last week, the physics meeting and an ad-hoc meeting to discuss issues he raised. Action items were formulated at the ad-hoc meeting. See Kei's presentation at this meeting for details.
The situation is still confused. A lot of the discussion was hampered by not having key experts in the room. We all know what we want, but there seems to be some history that needs unraveling before we get it. Those interested are referred to Kei's talks, the ad-hoc meeting page, and discussion on the email list. Mark intends to call another ad-hoc meeting soon.
b1pi and Single Track Test Failures
Last week Justin fixed a problem with inconsistency of the CDC code with recent changes from Richard to clean up the detector geometry. That fixed a class of failures, but we are still not getting successful completion of the b1pi and single-track tests. Paul and Simon have both seen the likely culprit. The symptom is an event that never completes. In addition, the problem is not reproducible. Simon is looking at it.
Data Challenge 2
- Mark has fired-up the old data challenge 1 jobs on the JLab farm, with the identical software, to re-establish the base for improvements for DC2.
- We would like to include beam-related electromagnetic background in DC2, among other things, but some study is necessary; the CPU costs may be too high. A volunteer to work on this is needed.