IPP Progress Report for the week 2011.05.30 - 2011.06.03

(Up to IPP Progress Reports)

Eugene Magnier

I spent part of this week working on PS1 science topics (yes, really!). I pushed the Megacam u-band data through the system, addressing a few megacam specific issues (mostly related to the high CR rate). I now have stacks for nearly all MD fields. I need to tie down the calibrations and then they can be used for science. I also have been working on the blue L dwarfs and L subdwarfs, with the intent of pulling together a paper.

In addition, I added some test points to pantasks and pcontrol so we can track down the causes of sluggishness and occasional crashes of those two systems. These seem to be related to memory or to jobs and or tasks that leak through the system. I also worked with Chris to fix the background error bug reported by Nigel for psphot analysis of the 3pi stacks. Finally, I worked with Michael Wood-Vasey who tested the ppSub modifications needed to run forward and backward convolution tests to choose the convolution direction. This fix will be pushed into the ops tag on our next release, aimed for the end of the week of Jun 6.

Serge Chastel

vacation

Heather Flewelling

vacation

Roy Henderson

Only worked 1.5 days this week.

  • PSVO
    • some time helping Kent Wood with his queries about PSVO/PSI
    • removed need for user to select a schema: this is evident from the database choice
    • implemented a plug-in query architecture for PSVO:
      • plug-in queries are defined in an XML file (xsd schema to follow)
      • plug-in can have parameters from which a GUI is automatically generated for the user to input values
  • PS1SC mailing list admin: some time spent trimming bouncing email addresses from the database

Bill Sweeney

  • studied Ohana implementation in preparation for moving *extract code to libdvo for re-use in relastro -high-speed
  • reviewed home directory usage and cleaned up obsolete data to recover space
  • holiday, caught up on email from holidays.
  • two days processing czar
  • monitored sts.2010 re-processing. Started sts.2009 reprocessing

Chris Waters

  • Vacation for two days
  • Registration: bugfix to use the same date (summitExp.dateobs) in both ippTools and ippScripts to determine if we should burntool an exposure or not. This should prevent hangups when this date and rawExp.dateobs disagree.
  • LAP: bugfix to allow lapRuns to share exposures better. This should result in a dramatic speed up in processing, as we will no longer need to reprocess an exposure multiple times. Fixed stuck lapRuns.
  • Diskspace: queued cleanup for old data that had not been destreaked previously. Rewrote ROC correction code and fixed a bug that allowed ROC compressed data to be shuffled across the cluster.
  • Stacking: more tests on stacking code, with evidence that filtering the pixels using KMM clustering should result in a more robust rejection algorithm.