OWG Meeting 18-Aug-2010

From GlueXWiki
Jump to: navigation, search



1:30 PM Wed 18-Aug-2010 CC F326


Next Meeting

1:30 PM Wed 1-Sep-2010 CC F326

New Action Items from this Meeting


Attendees: Elliott W, Andrew B, Justin B, Simon T, Mark I, Beni Z, Bryan M, David L, Yi Q, Hovanes E, Vardan G.

Event Display

Andrew summarized the summer's work on ded, the Hall D event display. See his talk for more details. He will continue to work on ded this academic year.

  • Java source is currently checked into the CLAS12 svn repository, will be moved to ours when bCNU is released.
  • How-to exists on installation, where to find source, etc. Prelimiary user guide created as well.
  • Scripts provided to start up with single command.
  • Code is fully Javadoc'd.
  • Currently need Eclipse to compile, will get Ant file working soon.
  • Input is evio file or evio formatted data on a socket. Will accept events from CLARA event source soon.
  • Geometry currently hard-coded in, not clear if it makes sense to get it from H-D XML geometry file.
  • About 10k LOC so far.
  • Might get another CNU student this fall to work with Andrew.
  • Future (see talk): 3-D, CLARA, improve B field, etc.


Justin summarized the summer's work on RootSpy. He will continue for one more semester. See his talk for more details.

  • Hist servers automatically discovered, can select hist by name or by server.
  • Multiple servers allowed on same node.
  • Full Root directory hierarchy supported.
  • Pull system, multiple displayers allowed.
  • Soon will add support for Root trees and on-the-fly user-supplied cut strings.
  • Hoping to deploy in upcoming Primex run.


Elliott reported on recent improvements to DANAEVIO.

  • Added about 15 more DANA objects to EVIO output file, one more (brand new) to go.
  • Working on making DANAEVIO into a CLARA event service.
  • Small number of EVIO banks can be read back in via EVIODANA plugin (from Dave L), this needs to be expanded.
  • Future EVIO improvements will unify file, buffer, cMsg and stream I/O.

Raw Data Format

We began discussing what the Hall D raw event format will look like. Many questions remained unanswered. We assumed that the DAQ group will take care of writing events to disk.

  • When do we need simulated raw events?
  • What will the event format look like? What kind of blocking scheme will be used?
  • We assume events will be fully built by the time they are written to disk by the ER. Is this correct?
  • The data format coming out of the front end modules is known. How this is transformed by the event building process is unknown (to us).
  • We want to read out all ADC samples periodically to check the summing procedure. Not clear how this will be implemented.
  • Can the front-end modules be programmed to periodically send all samples in high-speed mode?
  • Probably want all samples from all crates and modules in a single, very large event.

Coil Test

  • Magnet, coil, yokes, spacers, stand and platform are all installed.
  • Cryo can will be installed by the end of Aug.
  • Hope to go cold by mid to late Sep.
  • PLC programming proceeding well.
  • HMI screens and archive system being developed by Yi.