GlueX Offline Meeting, July 27, 2010

From GlueXWiki
Jump to: navigation, search

GlueX Offline Meeting
Tuesday, July 27, 2010
2:00 pm EDT
JLab: CEBAF Center F326/7

Telephone

To connect by telephone:

  1. dial:
    • 800-377-8846 : US
    • 888-276-7715 : Canada
    • 302-709-8424 : International
  2. enter participant code: 77438230# (remember the "#")

Video Conferencing

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2010-07 on the JLab CUE (you have to be a member of the "halld" Unix group). This directory is accessible from the web at https://halldweb.jlab.org/talks/2010-07/ .

Agenda

  1. Announcements
    1. SIMD enabled code disabled: ?
    2. Make "docs" area of subversion repository require authentication on checkout?: Mark
  2. Review of minutes from the last meeting: all
  3. Event Display: Andrew
  4. Generalized particle gun: Mark
  5. Documentation Policy: Mark
  6. GlueX shell environment set-up policy: Mark
  7. Action Item Review: all

Minutes

  • CMU: Curtis Meyer
  • JLab: Andrew Blackburn, Hovanes Egiyan, Mark Ito (chair), Yi Qiang, Sasha Somov, Simon Taylor, Elliott Wolin, Beni Zihlmann

Announcements

  • SIMD-enabled code disabled: But only for 32-bit Linux machines. The problem seem to be a bug in version 4.1.2 gcc compiler which is the default for the Lab's desktop Linux boxes. The code does not have a problem on any other platform.
  • Make "docs" area of subversion repository require authentication on checkout: We are not going to do this.

Review of minutes from the last meeting

We reviewed the minutes of the July 13 Offline meeting.

  • OSG Toolkit: Mark mentioned that Richard Jones has written the documentation on the OSG client software that was requested by IT Division before opening the necessary network ports at the Lab.
  • Software Workshop: David reported that the proposal is due to JSA next Friday. It has been circulated among the committee and the final draft is waiting on some numbers from Ed Brash.
  • 48-core machine: It is gone now. David ran some stuff, and captured some performance counters. There is a lot of data there, next step: analysis.
  • Another plug-in trained collaborator: Working from David's talk from last week, Beni was able to write one of his own. We still need a HOWTO on the subject.

Event Display

Andrew gave us a demo of the new event display. It is pure Java and is constructed within Dave Heddle's bCNU framework. See his HOWTO for instructions on how to get and run it.

He showed us a few events. The code reproduced the views available in hdview2 and adds another for the BCAL.The framework provides many useful features, such as a heads up display, zooming, multiple views in a single frame, etc. Currently, the geometry is hard-coded into the routines.

Lots of feedback was offered on a variety of features.

We veered briefly into a discussion of HDDM vs. EVIO (as we often do). We will have to discuss our attitude toward these (and other for that matter) formats soon.

Generalized particle gun

Mark walked us through his recent email announcing a simple perl-based approach to a script driven single particle gun. Although this functionality is already present in HDGeant, having all particle kinematics under script control allows for arbitrary variation from event to event, e. g., the event vertex can be moved under script control. The output is HDDM-conforming xml and thus can be converted to an HDDM file for input to hdgeant, the conversion done with existing tools.

Documentation Policy

Mark reviewed a wiki page, from last March, outlining a structure for our documentation and suggested that we adopt it officially. Although the content is nothing more than common sense, having an agreed upon structure will help frame documentation discussion/decisions in the future. One component that needs to be introduced is a GlueX Offline FAQ. Mark agreed to start on this.

No objections were offered and the suggestion was adopted.

GlueX shell environment set-up policy

There are two methods that exist to set-up the GlueX environment for individuals. One (setenv.csh) captures a working environment and allows one to reproduce it faithfully in the future. The other (gluex_env.csh) respects key choices made in the current environment, supplies default values for those not defined, and defines others based on the key variables. The former is useful for reproducing results, the latter for first-time environment set-up or non-standard configurations. As such Mark thought that in this case we should support both approaches (not much work really), and document each. No objections were raised.

Action Item Review

We reviewed recently resolved issues and those outstanding.

New Action Items

  1. Write a HOWTO on plug-in's in JANA -> David
  2. Schedule an HDDM/EVIO discussion -> Mark
  3. Write the GlueX Offline FAQ -> Mark