Core Minutes 3/23/2004  

Present:   Sandhia Bansal, Ursula Berthon, Joanne Bogart,  Toby Burnett,  David Chamont, Seth Digel, Richard Dubois,  Berrie Giebels, Navid Golpayegani, Heather Kelly, Traudl Hansl-Kozanecka,  Matt Langston, Pat Nolan, James Peachey, Sean Robinson, Leon Rochester, Alex Schlessinger, Tom Stephens, Tracy Usher

• Core issues and status: (Toby) See Toby's  run-down of issues of interest here and follow-up on some of them below. 

  • Release Manager (Alex): When attempting to run a test program from one package, it sometimes has trouble finding a shared library belonging to another even though, as far as Alex can tell, the library has been properly built. He's been cleaning up after such failed builds by hand, but this needs to be understood. Toby suggested that, since the Release Manager is a crucial component that needs to be up full-time, Alex delegate someone in a distant time zone to keep an eye on it when he can't. 

    In other RM news, Alex reported that Windows builds succeed on the box in his office.  There was a problem with a missing formally-required, even though unnecessary, registry entry which he has addressed. He is engaged on the last bit of work needed before Windows builds can be done routinely by Release Manager: soemthing to smooth over the differences between submission to LSF and to a Windows server.

  • ROOT conversion service (Heather, Ursula) Drop-dead date for completion is end of May; code review will then probably be scheduled in June. In addition to some clean-up, code needs to be added to handle relations.
  • Kalman filter (Toby) He made some plots which clearly demonstrate the suspected dependence of the chi squared calculation on phi.  Work on understanding the reason is ongoing.
  • Parameter files (James) His presentation (nicely self-contained; no point in reprising here) described typical use of pfiles among astronomers and how this impacts GLAST ScienceTools usage.

    One "feature" of pfiles implementations is that a user file will get clobbered if the corresponding system file is newer.  Chances are the tool using the file has been upgraded and the user file is no longer appropriate.  Toby suggested that, in this case, the system first make a back-up of the user file.

    So far it looks like pfiles and the Install Area are compatible. There is a requirement for unique filenames, but it causes no particular burden.

  • ScienceTools release, reoranization (Toby) Newly-released v2r0 builds with VC 7 and 7.1, gcc 3.2.2, but not so far with gcc 3.3 or 3.2.3. Various packages have been temporarily removed from LATEST while they are converted to use the as-yet-unnamed Goodi replacement.

• New blood: (Berrie) David Chamont, coming with much experience from CMS, is joining us.  Projects he's been involved in include ORCA and OVAL (which, Richard pointed out, bears some relation to our own Release Manager).

• Fuzzy clusters for Windows: (Navid) The port has been hung up by the use of the complex class in CLHEP; Gabriel is writing a wrapper for Windows.

• System tests: (Matt) This schedule shows how the work is broken up.  Dates and assigned names should be taken with a grain of salt. Concentration right now is on items 71-76, concerning work on a ROOT class library, involving a new (for us) paradigm. Review by the ROOT-savvy is encouraged (more than encouraged, in Heather's case).

• Windows boxes at SLAC: (Richard) The original plan (2 all-purpose boxes) has been modified somewhat since the Computer Center doesn't want that kind of mix.  Instead we'll have one box set up as a Citrix server (for interactive running of the Windows builds or Windows-specific applications) and another two for Windows Release Manager building and for System Tests. Work on this is temporarily stalled since the person assigned is ill, but it should be back on track soon.

• ScienceTools walk-through teaser: (James) Featured topics will be the new hoops and the new data access component (for the time being to be found in users/peachey/table).

• Flight Integration: (Richard) The promised weekly run-down..

  • Opus: (Matt for Dan) Matt believes Dan's work on stored procedures in on track, with completion expected at the end of the week.
  • Calibration database interface: (Joanne) Some progress on underpinnings, but still plenty to go.
  • Tracker report: (Leon) Had a discussion with Eduardo on what tests will be needed.
  • Analysis support: (Richard) Probably will want to define a new n-tuple and a procedure to regenerate easily. Runs will be done in a large variety of configurations and we have to keep all this straight. Anticipate that System Tests will be primary presentation medium for the data; it will have to do adequate book-keeping for the different configurations.

  Action Items 

• Document effect of MC pruning options (Riccardo) (8/5/2003) It's documented in the code; Tracy will look into putting the information in a more accessible place. (9/2/2003)
• Write up, and perhaps present at a future meeting, hints on debugging. (02/18/03)   Some combination of Toby and Tracy will provide something for Windows.(7/22/03) 
• Overhaul documentation altogether by about November (Heather and ??) (9/2/2003) In progress; not yet done (01/13/04)
• Resolve calibration/timestamp issue. (11/20/02) Online folk have not yet thought about putting timestamp information into EM data, but, thanks to Richard's inquiry, may now do so. (2/25/03) We can probably live without a realistic system for EM, but we will put it on the agenda for the DC1 workshop.(5/6/03) At the DC1 Workshop Steve Ritz  promised to find an authoritative reference for timestamp format and pass it on. (7/17/03) Now that events have "real" timestamps, Joanne needs to reassess situation with Calibration (9/2/2003) Timestamp exists in EM data. Needs to be fetched by Calibration (1/13/04) Put on hold until formats settle down. (3/16/04)

previous

minutes index

next

J. Bogart Last Modified: 01-Jun-2010 15:47:59 -0700