GlueX Offline Meeting, September 4, 2013
GlueX Offline Software Meeting
Wednesday, September 4, 2013
1:30 pm EDT
JLab: CEBAF Center F326/327
- Review of minutes from the last meeting: all
- Report from Work Flow meeting (Mark)
- CDC Track Finding (Paul)
- October Collaboration Meeting
- 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/halldweb/html/talks/2013-3Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2013-3Q/ .
- CMU: Will Levine, Paul Mattione, Curtis Meyer
- IU: Kei Moriya, Matt Shepherd
- JLab: Mark Dalton, David Lawrence, Mark Ito (chair), Simon Taylor, Elliott Wolin
- MIT: Justin Stevens
- Northwestern: Sean Dobbs
- UConn: Alex Barnes
Review of Minutes from Last Meeting
We looked at the minutes from the August 21st meeting. No major comments.
Report from Work Flow Meeting
Mark reported that last week there was another meeting of this committee. Mainly it was information gathering by Scientific Computing staff. They had questions about what practices for work flow management are being used or planned for Halls B and D. There will be another meeting in three weeks.
CDC Track Finding
Paul has done an almost complete rewrite of the CDC track finding code in order to support the goal of elimination of ghost tracks from curling tracks in the CDC. That having been said, the resulting code can be used for all tracks with all or part of their hits in CDC. He gave a detailed description of the ideas and some comparative studies of the results (existing code vs. his new "spiral" code). See his wiki page for his presentation. Some highlights:
- The new code finds clusters of hits ("super-layer seeds") in both axial and stereo layers. The existing code only does so for the axial layers.
- Tracks are found from the inner super layers to outer, using stereo layers and always adding hits from a complete seed.
- Note is made of seeds that are consistent with a track that is at the outside of a curling track. These have a pattern that is roughly aligned with consecutive straws in a layer rather than roughly along a radial line in the CDC. This information is used later to distinguish between track candidates that are actually pieces of the same curling track.
- He has studied performance for single tracks and for a variety of physics topologies.
- He has developed a new plug-in that refines the efficiency calculation used in the bi-weekly single-track reconstruction tests. The new method requires that a found candidate or time-based track agree with the generated parameters to be considered found.
- Overall he sees improvement in track finding efficiency, especially at low momentum at polar angles greater than about 30 degrees. Improvements can be a large as 10% in (εnew−εoldεold. Places where efficiency is degraded are few (if there are any at all).
- In many cases, although track finding efficiency is improved relative to the existing code, the existing code recovers some efficiency at the track fitting stage (using the Kalman fitter).
Again interested parties should look at his wiki page to get the details and quantitative characterization of the improvements.