IPP Progress Report for the week YYYY.MM.DD - YYYY.MM.DD

(Up to IPP Progress Reports)

Eugene Magnier

  • NSF proposals

Serge Chastel

  • Czar on Election Day
  • More data for mops tests
  • Still working with condor: hosts can be "properly" targeted.
  • Configuration fix: Diffs are now all in PS1_DV3 format

Heather Flewelling

  • validation of SAS-10 IPP->PSPS ingest

Mark Huber

  • MD03 refstack finished and backlog of diffims done
  • MD05 continued setup and reprocessing of exposures, will also be finally finished shortly
  • little time on continued investigation of WS diffim PSF mag offset

Bill Sweeney

  • tested new web interface for random access to smf files. A couple of minor changes are needed before we can open this up to PS1SC.
  • Investigated ways to reduce the amount of redundant storage used for single frame detections.
  • started cleanup of 2010-2011 camera stage distribution bundles from the data store
  • two time consuming days as processing czar including cleaning up after a crash of postage stamp server/data store machine ippc17
  • updated psLib's table of leap seconds to include new one added July 1 of this year. Spent some time testing the fix.

Chris Waters

  • Updated configuration for SweetSpot? diffs to use dual convolution for warp-warp, and SINGLE2 for warp-stack. This seems to provide better results as the convolution direction is generally correct now.
  • Ran test with a sample of images already processed by MOPS to investigate the behavior with the new convolution. Results available at http://svn.pan-starrs.ifa.hawaii.edu/trac/ipp/wiki/MOPS_diff_false_positives
  • Finished background stacking/projection cell stage products code for those processed along with the initial processing. This doesn't allow previously processed data to have these products generated at this time, but I've worked out pswarp recipes that allow for the creation of warped background models for cleaned data. The only remaining step is to finish a ppStackMedian command to allow these warped background models to be directly combined in parallel to a standard processing run. This program will also allow the creation of the simple median stacks that have been requested.