[Halld-offline] Software Meeting Minutes, April 16, 2019
marki at jlab.org
Fri Apr 19 14:53:56 EDT 2019
Please find the minutes here
Minutes of the April 16, 2019 GlueX Software Meeting
* *CMU: * Naomi Jarvis
* *FIU: * Rupesh Dotel
* *JLab: * Shankar Adhikari, Alexander Austregesilo, Sean Dobbs,
Hovanes Egiyan, Stuart Fegan, Mark Ito (chair), David Lawrence,
Justin Stevens, Simon Taylor, Beni Zihlmann
There is a recording of this meeting <https://bluejeans.com/s/qT@i7/> on
the BlueJeans site. Use your JLab credentials to access it.
David sent around an email
asking about interest in having the lab be a satellite site for an HPC
Summer Boot Camp that teaches fundamentals in parallel programming to be
held June 3-6. If you have an opinion, please contact David by April 17.
Review of minutes from the March 5 meeting
We went over the minutes
* David reported that Batch 3 of the reconstruction of Spring 2018
data is complete. Batch 4 was submitted to NERSC earlier today.
* David, Alex, Curtis Meyer, and Amber Boehnlein submitted a proposal
to the Pittsburgh Super Computer Center for enough time to do a
single reconstruction pass on the Fall 2018 data (7.65 SU's).
Requests take about two weeks to be evaluated.
Report from the last HDGeant4 Meeting
We went over the minutes from the meeting on April 9
We though that Peter might want to pick a few items from his comparisons
between G3 and G4 and post them as issues on the HDGeant4 page on GitHub.
Naomi brought up a couple of separate issues with HDGeant4:
1. The number of events to be generated can be specified in the macro
that starts the job or in the control.in file. We don't which is
controlling. David mentioned that he has requested that it be
specify-able in control.in to maintain symmetry between hdgeant and
HDGeant4. We again suggested that this was an appropriate issue for
the HDGeant4 GitHub site.
2. The "-r" switch for setting the run number does not work.
Mark described recent personnel changes
in IT Division.
PrimEx branches in HDDS
There is a new system for maintaining alternate geometries in the Git
repository. Richard and Sasha described it in a recent email exchange
Recall that the geometry for reconstruction should be coming from the
CCDB and not the Git repository. The versions in the repository are only
for developing new geometry configurations. In the repository it is
important that distinct geometries be kept on separate branches.
"Merging to the master branch" is a disaster in this case.
Review of recent issues and pull requests
We reviewed issues for halld_recon and halld_sim. We discussed how to
prioritize work on the various issues. In real time, Sean assigned
color-coded labels <https://github.com/JeffersonLab/halld_recon/issues>
to all of the extant halld_recon issues. Issue #71, Crash in ST matching
<https://github.com/JeffersonLab/halld_recon/issues/71> is marked as the
highest priority issue, as we agreed.
General Documentation in Need of Update
Hovanes remarked that the general documentation for the GlueX software
on the wiki page is inconsistent and out of date. No one disputed his
Sean and Mark will meet with Dmitry to plan work to address the issues
reported with CCDB.
* This page was last modified on 19 April 2019, at 14:49.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Halld-offline