Teraminutes of December 15th 2006

Present: Chiara, Yannick, Emmanuel, Fred, Mireille, Henry.

Absent: Laurent, Mathias, Anthony.

Emmanuel's summary document

Megacam PI data (Yannick, Mireille, Emmanuel)
 04A26 (Soucail); Yannick will validate the data over the weekend.
 Annie Robin data; this is a very crowded field, and there seems to be still some problems of matching: Mireille will send the groups which don't work to Emmanuel for further investigations.
 Mireille also noticed that the selection of the photometric reference frame isn't stable. Emmanuel will investigate.
 Emmanuel also mentions that 2MASS seems to almost always give the better astrometric solutions.
 Yannick mentions that CFHT has found a drop of 12% in the photometric zero point over the last six months. Jean-Charles has asked us to look for this effect in our data.
 New filter curves for MEGACAM recalculated by Mathias and the folks from SNLS: Henry will send them to Roser Pello and Olivier Ilbert.
 Mathias should participate more to the processing of Annie Robin data, in particular.
 Yannick will send out a mail to J.J. Kavelaars to ask him to download the W1-u extended "Willott" data and to be informed about the conclusions of the last CADC meeting concerning the archiving of "special" data.
 Fred should transfer the UH Capak data, Laurent should prepare the jobs, and Mireille should follow the processing.

Hardware/System
 Power failure on some TERAPIX lines: last night the breakers tripped again, 40A fuses burnt, without any apparent reasons. Replaced in the afternoon.
 The migration of clix and associated machines didn't happen because of these problems.
 Mix9 still causing problems on Saturday morning. Fred and Chiara will check if the newer version of the kernel installed by Fred fixes the probleme tonight.
 48 Port gigabit Netgear switch has arrived.
 As of Friday evening, mix3, mix4, clix and dbterapix were still to be upgraded to the new version of the system. Replication to mix7 is on hold until one is certain that the new system is able to handle perfectly the pipeline workload.
 Emmanuel will discuss with Fred early next week about user access and software installation on the cluster for Olivier Ricou's work on webservices.
 It's been proposed to back-up CFHTLS-T03 metadata at CCPN in Lyon.

Software development
 Chiara tried to run SPICA with her new version of Weightwatcher on pix3. Unfortunately SPICA appears to be broken for WIRCAM data (problem to access flat fields).
 The bug in SCAMP with number of exposures >256 fixed. More tests needed (photometry, escalope) needed before official release.
 The PSF convolution bug in SkyMaker will be fixed ASAP.
 Emmanuel's galaxy profile fitting program works at around 3 galaxies per second; he would like to have this for T04 but it's not certain it will be ready in time.

T04 release
 Fred needs to identify and remove all exposures in double among the 2000+ that were downloaded from CADC before Laurent can ingest the data in the database.
 Laurent will be on holiday starting from December the 19th (until January 9). If everything is ready by then, Chiara may start the QFITSin.

Other matters
 Fred has responded to the folks from Naples, and proposed some dates, possibly in January.
 UltraVista and KIDSS Viking data are likely to arrive at TERAPIX in mid-2008.
 Henry will put the UltraVista proposal on the TERAPIX intranet.
 Emmanuel will become a member of the Dark Energy Survey team (DES), as they are using his software (sextractor, scamp, swarp). A visit is planned to NCSA, Illinois next summer.
 Webservice VO image simulation meeting yesterday morning. Potential problem outlined by Emmanuel and Yannick concerning with "software visibility" in Grid applications (i.e. TeraGrid).
 It is hard to find a stagiaire to work on this stuff because of the new internet boom; Olivier at LRDE is ready to work on webservices but he will need access to the machines; some discussion still needed with Fred on Monday about authorizations and software installation.
 Yannick will post the TERAPIX IE job offer on the AAS boards
 Yannick will answer to Ed Valentijn about the AstroWISE system installation and the Panorapix position at the beginning of next week.
 Emmanuel will send to Ed Valentijn a short report about the work on OpenGL in TERAPIX early next week.
 Pending tasks in Dotproject will be reviewed next Monday.

WIRCAM
 The CFHT board is currently meeting, and hopefully they will discuss the problems with lack of manpower at CFHT for WIRCAM reductions.
 Chiara will use the raids of mix1 for Wircam processing; no-one objects.

Next meeting: Thursday 21 December 2006 at 9:30 am