GlueX Offline Meeting, June 24, 2015

From GlueXWiki
Revision as of 13:00, 5 October 2015 by Marki (Talk | contribs) (More on Git for GlueX)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

GlueX Offline Software Meeting
Wednesday, June 24, 2015
1:30 pm EDT
JLab: CEBAF Center F326/327

Agenda

  1. Announcements
    1. nightly build switched to jana 0.7.3
    2. new volatile disk hardware for halld
    3. Farm outage begins Friday June 26 4pm with /home upgrade
    4. JLab Scientific Computing - power outage June 27
  2. Offline-style builds on the Gluon cluster (Sascha)
  3. Review of minutes from June 10 (all)
  4. Offline Monitoring (Kei et al.)
  5. Spring 2015 Commissioning Simulations (Sean)
  6. More on Git for GlueX (Mark)
  7. Action Item Review

Communication Information

Remote Connection

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2015 on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2015/ .

Minutes

Present:

  • FSU: Aristeidis Tsaris
  • JLab: Mark Dalton, Mark Ito (chair), Mahmoud Kamel, Paul Mattione, Kei Moriya, Eric Pooser, Matt Shepherd, Sascha Somov, Nathan Sparks, Justin Stevens, Simon Taylor, Beni Zihlmann
  • NU: Sean Dobbs

Announcements

  1. Nightly build switched to JANA 0.7.3
  2. new volatile disk hardware for halld
  3. Farm outage begins Friday June 26 4pm with /home upgrade
  4. JLab Scientific Computing - power outage June 27

Review of minutes from June 10

We reviewed the minutes from last time.

Paul is working on updates to the Analysis Workshop we had in July of 2013. Much has changed and the material needs a refresh. He will present part of his work at the Physics Meeting on Monday. At that meeting Justin will also show some plots of signals from the commissioning runs, identify areas for analysis work, and identify available plug-ins.

Offline-style builds on the Gluon cluster

Sascha proposed that we have offline-style builds on the Gluon Cluster so that collaborators can have uniform structure to deal with. He pointed out that the cluster is convenient to use for newly-taken data. We had some discussion but at some point realized that all of the offline builds are already available on the cluster via the group and scratch disks.

In the course of the discussion, a need for better documentation was identified, documentation for using the various offline builds (tagged releases, nightly builds, etc.). Mark I. will look into getting the information out there.

We also discussed adding a nightly build on Beni's Ubuntu system. Beni and Mark I. will discuss this.

Related to our builds, Matt mentioned a common pitfall using a mixture publicly and privately built software: inconsistent header files. This one can be very hard to diagnose. He noted that on CLEO, the include directories[?] contained a ".release" file that gave version information. The build system would check these files and give an error if inclusion of inconsistent versions were attempted. Something to consider.

Offline Monitoring

Kei presented slides on the latest status and news. His outline:

  • Status of current launch
  • Errors reported from SciComp in offline monitoring jobs
  • Relating event numbers and real time
  • SWIF

Please see his slides for further information.

Processing Update

Sean described recent improvements to the system for processing the results of the offline monitoring jobs. See his slides for a description of the work.

Spring 2015 Commissioning Simulations

Sean described the status.

  • We are basically ready to run.
  • The new BCAL simulation is ready. The shower energy is a few percent low, but that was the case with the previous version of the code.
  • There are also improvement in tracking that Simon has checked in.
  • Mark will tag a new version of sim-recon for this effort.

More on Git for GlueX

Mark presented further thoughts on how we might conversion from Subversion to Git for source code version control. See his slides for the details.

Mark proposed that we make a decision on the switch at the next offline meeting.