GlueX Offline Meeting, July 26, 2017
GlueX Offline Software Meeting
Wednesday, July 26, 2017
11:00 am EDT
JLab: CEBAF Center F326/327
- Review of minutes from the last meeting (all)
- HDvis update (Dmitry, Thomas)
- 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
- FIU: Mahmoud Kamel
- JLab: Alexander Austregesilo, Thomas Britton, Eugene Chudakov, Hovanes Egiyan, Mark Ito (chair), David Lawrence, Dmitry Romanov, Justin Stevens, Simon Taylor
- MIT: Cris Fanelli
There is a recording of this meeting on the BlueJeans site. Use your JLab credentials to access it.
- Major IT Outage Set for Saturday, July 29, as announced in the Weekly Briefs. Central systems will be relocated from the second floor of CEBAF Center to the Data Center.
- Added capabilities in OSG run scripts. Improvements to the grid submit mechanism from Richard Jones. There was subsequent discussion on the email list of further suggestions from Sean Dobbs.
- Reconstruction Launch: 2017-01 ver01. Alex reviewed items from his email.
- Singularity 2.3.1 install for the farm on CentOS7 nodes.
- New sim-recon release: version 2.16.0 and version_2.13.2.xml and new HDPM release. Multiple packages had version upgrades.
Review of minutes from the last meeting
We looked at minutes from the July 12 meeting.
- CCDB development. Dmitry has fixed the crash-at-end-of-job problem. It is incorporated in the latest CCDB release (1.06.05). He has started on work modernizing the code to use C++11-era smart pointers.
- Mini-launches started. Justin reminded us that Sean has started simulations for mcsmear development.
Dmitry described the current architecture and how we got here. The original attempt at a 3-D event display used the EVE package of ROOT. Problems EVE and in integration with JANA were many and varied. The current scheme uses three.js as the underlying rendering engine with events being fed to the browser-based display from a JANA server or read directly from a JSON-formatted disk file. Three.js underlies a wide variety of web applications (including a STAR event display) and has a large user base. See Dmitry's slides for architectural diagrams and illustration of other three.js applications.
Thomas demonstrated the current version of HDvis using an example event. Time evolution of events has been implemented, i. e., an animated view with an event time clock, where charged tracks are drawn propagating according the event clock, and detector hits "light up" at their measured time on the event clock. Further verbal description cannot do justice to the demo, which starts at the 38:30 mark in the recording.
Mark identified two labor-intensive, on-going tasks as described to him by Dmitry and Thomas:
- translation of JANA objects to JSON format
- extraction of geometry information from HDDS
There were a lot of good questions, discussion, and suggestions covering a wide variety of topics, overwhelming the secretary. Those that contributed are encouraged to restate feedback to the development team via any convenient channel, including the issues page on GitHub.
As a bonus, virtual reality was trivially implemented. Those at JLab were treated to the 3-D VR experience of the event using Thomas's Oculus Rift. This feature elicited an OMG from Pittsburgh.
Curtis reported that he had contacted GSI about taking on HDPM maintenance. They discussed it and turned down the offer. Other avenues are being pursued.