Core Minutes 6 March 2007

Core Minutes 3/6/2007

Present: Joanne Bogart, Anders Borgland, Toby Burnett, Jim Chiang, Johann Cohen-Tanugi, Richard Dubois, Warren Focke, Berrie Giebels, Heather Kelly, Michael Kuss, Bryson Lee, Tom Glanzman, Navid Golpayegani, Chuck Patterson, Igor Pavlin, Leon Rochester, Tracy Usher, Larry Wai

Disk (lack of): (Richard) The Computer Center doesn't have adequate power for everybody's new hardware: ours, Atlas, etc. Richard's suggestion that Babar could do with a bit less without noticeable impact did not meet with enthusiasm. Another possibility would be to replace old, slow hardware with newer models which use much less power and do more. Meanwhile Richard will investigate and document various Draconian alternatives.
New OS support:
- RM and RHEL4 (Navid) is still battling a problem with a double free in the finalize step of the AcdUtil test program.
- Windows upgrade status (Toby) Is lacking a proper compiled version of Gaudi. What is the procedure for creating it? (Navid) uses the CMT requirements that it comes with. Heather wrote up some instructions once.
ROOT news: (Richard) There is now an official new tag of 5.14 incorporating the fix we need. (Toby) will see about getting a Windows build for it.
(Richard, acting as mouthpiece for those who can only listen at the meeting) Tracy is waiting for Heather's seal of approval on his ROOT-related changes to several packages. Heather can't yet get the mcRootData test program to run.
NTuple additions (Leon) will get together with Toby when he is at SLAC later this week. (Toby) would like to add a use of astro since some of the necessary information resides there. (Leon) Maybe that's ok; he isn't quite ready to sign off on it yet. (Richard) Once the variables are in, will make a 2 or 3 billion event background run for Bill's rejection analysis. Also need input for Interleave with slightly different requirements: for Interleave we segregate out the earth albedo and run it in zenith angle bins. These requirements may be mutually exclusive, in which case we'll need a separate run for Interleave.
obsSim: (Richard) expects an obsSim run soon. Alice Harding is shepherd; Nicola is working on code bits, pipeline. (Tom) has been communicating with Nicola, who is asking good questions; it looks like progress. By the way, what about the Sky Model pulsar problem ? (Richard) No known resolution yet. [later: Max (Razzano) says the problem has vanished with recent code updates]
SAA hang: When the LAT is within the SAA for a complete run, so that no events are generated, the run hangs, only exiting when it reaches the queue time limit. This problem first appeared with GR v9r22. (Toby) will look into it if he is supplied with information needed to reproduce it.
Announcements: (Richard)
- End-to-end test (ETE1) is scheduled for the 17th of this month. Data will flow from the LAT, to an antenna outside in the parking lot, to TDRS, around and around, and ultimately to SLAC. Plan to process it with Pipeline 2. There will be a pre-test tomorrow of L1 processing using input from Tom's L0 simulations.
- Heather returns next week. As Infrastructure Queen, she might logically take over running of these (Core Infrastructure) meetings.

minutes index

J. Bogart, Last Modified: 25-Jul-2007 11:25:57 -0700