GlueX Offline Meeting, August 23, 2017
GlueX Offline Software Meeting
Wednesday, August 23, 2017
11:00 am EDT
JLab: CEBAF Center F326/327
- 1 Agenda
- 2 Communication Information
- 3 Minutes
- New version of RCDB: version 0.02
- New version of mcsmear development branch
- Pull request tests not working
- Review of minutes from the last meeting (all)
- Status of HDvis (Dmitry, Thomas) current state
- Simulations for mcsmear development (Sean)
- Review of recent pull requests (all)
- Review of recent discussion on the GlueX Software Help List (all)
- Action Item Review
- The BlueJeans meeting number is 968 592 007 .
- Join the Meeting via BlueJeans
Talks can be deposited in the directory
/group/halld/www/halldweb/html/talks/2017 on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2017/ .
- CMU: Curtis Meyer
- FSU : Brad Cannon, Sean Dobbs
- Glasgow : Peter Pauli
- JLab : Thomas Britton, Eugene Chudakov, Hovanes Egiyan, Mark Ito (chair), Dmitry Romanov, Justin Stevens, Simon Taylor, Beni Zihlmann
There is a recording of this meeting on the BlueJeans site. Use your JLab credentials to access it.
- New version of RCDB: version 0.02. This version was needed to support Sascha Somov new trigger simulation code. there was a mistake in the announcement: the new version is actually 0.02.
- New version of mcsmear development branch. Mark will announce a build of the latest tag on the mcsmear development branch before lunch today. [Announcement, added in press].
- Pull request tests not working. Sean reported that the problem has something to do with ssh from the webserver to the build node. He and Mark are working on it.
- CCDB 1.07 on the way. Dmitry will release this in a week or so. I includes better logging and log searching, an update of "smart" pointers to use those native to the C++11 standard, and eliminates the possibility of user "anonymous" contributing constants. Lately, Dmitry has been getting help from Andrew Lendacky, a Hall-B student from Duquesne University.
Review of minutes from the last meeting
We went over the minutes from August 9. There was no significant discussion.
Status of HDvis
Thomas gave an update including a demonstration of the latest version. For his commentary, see the 16:30 mark in the recording.
- The FDC hits are aligned properly now.
- CDC hits have been added.
- Start Counter hits have been added.
- Currently missing BCAL geometry and hits.
- There are thousands of objects in the scene at present; this is affecting performance.
- Dmitry has been working on the structure of the code, working toward a framework within which other developers can contribute visualization schemes.
- A play/pause button will be added.
- Keyboard for translations, mouse for rotations.
Simulations for mcsmear development
Sean made several comments on the recent work by folks to do comparison between Monte Carlo and data.
- For this work we should be using the recon-2017_01-ver01-batch01-mcsmear branch of sim-recon. This has the reconstruction code for the latest reconstruction pass on Spring 2017 data, but updated smearing code.
- The trigger simulation has been added by Sascha. Sean is looking at how to use it so that we all get a consistent trigger cut when requested.
- He is looking at HDDM changes.
- Adding trigger information to the event.
- Putting in pulse amplitudes. These are not there at present, but are used in a lot of contexts in the reconstruction.
- He has been testing the framework for generation of Monte Carlo on the OSG. He has been working off Richard's tools and has had some success, often getting better throughput than possible on the JLab farm.
- Looking at a scheme where we have a uniform interface to both the farm and the OSG, including better accounting. Keeping track of jobs is especially important on the grid where jobs can be pre-empted and may need to the re-submitted.
- Justin asked if SWIF could be adopted to use on the OSG. This has been discussed but is not on anyone's current work list.
Review of recent pull requests
- Beni has added ADC and TDC timing cuts to TOF point formation.
- David has added material maps for detectors downstream of the tracking chambers (TOF, FCAL).
- David has added a program to hd_utilities to reduce the size of SQLite filed, discarding data outside of a user-specified run range and variation. This can reduce the size of the file by an order of magnitude.
Review of recent discussion on the GlueX Software Help List
There has been an extended discussion of timing in the Monte Carlo centered around timing differences induced by an alternate location of the target in z, specifically, the target for the charged pion polarization experiment. This means relative to a clock based on the RF, the interaction time is different for the two target positions. Please see the list for the on-going discussion.
Action Item Review
- Release the tagged version of the mcsmear development branch. (Mark)
- Rationalize the version set associated with the new RCDB version. (Mark)
- Fix the pull-request-induced build mechanism. (Mark and Sean)
- Add the BCAL to HDvis (Thomas, Dmitry)
- Sort out timing issues related to target position as discussed on the Software Help list.