IPP Progress Report for the week 2011.06.13 - 2011.06.17

(Up to IPP Progress Reports)

Eugene Magnier

Serge Chastel

  • Sick on Thursday
  • Apache/nebulous load balancing
  • MOPS czar last weekend

Heather Flewelling

  • ThreePi? things
    • minidvodbs transferred to ipp022
    • rsyncing copies of ThreePi? dvodb on maui cluster
  • ifaps1 liason
    • sent out instructions on how to access ifaps1 to someone new
    • transferred latest SAS stacks to ifaps1 for TD.
  • isp
    • astrometry is as good as it gets (it fails when skyprobe is obscurred by the dome)
    • tried making the dark better
    • There are some problems with extra fake sources- I think it might need a static mask
  • grabbed files for JT
  • reprocessed comet for RJW
  • czar

Roy Henderson

  • PSVO: I've now implemented the two major enhancements outlined at Boston: plug-in queries and Derby Db
    • can now save any results to the Derby database, and reload in TOPCAT, or any other client
    • some re-factoring to use Starlink packages to represent tables
    • now using Starlink TableModel objects, instead of my own
    • all query tabs now close neatly, meaning they remove any temp files they may have used
    • a working dir is now created for storing the DerbyDB and any temp files
    • integrated Derby querying into normal query panel, i.e. you can query local Derby Db just like you can query PSPS, MOPS etc
  • more grand-reprocessed data for Jim to check
  • PS1SC wiki:
    • upload restriction issues
    • some discussion and planning for the big migration with Steph and Ken
  • meeting to discuss outstanding stack issues for loading

Bill Sweeney

  • Selected exposures for and built a reference stack for the new STS fields. 250 exposures with fwhm_major < 1 arc second where chosen.
  • changed stacktool -definebyquery to accept multiple -skycell options so that stacks for a subset of a collection of skycells may be easily queued.
  • spent some time investigating why some sts stacks failed. The skycells were all in the very corners of the observations and the inputs did not have enough pixels for ppStack to process. (We require >= 10% good pixels. These input warps had less than 3%).
  • Performed a magic test using difference runs using the new stack as a template. The results were abysmal. Many more false streaks were detected and the masking was much higher than when warp-warp diffs were used. We are going to need to adapt the IPP to handle 3 exposures per night.
  • Finished coding and testing the new relastro -high-speed and the associated code reorganisation. This task took much longer than we expected. Wrote a note describing how to use the new code and sent it to a couple of users for testing.
  • 1 1/2 days as processing czar. Due to lack of space on the wave3 nodes the wave1 nodes are getting lots of nfs activity and they often overload. Several jobs left corrupted files which needed to be fixed by hand.
  • Gathered up raw and processed images from the standard star observations for John Tonry to analyze.
  • debugged an ppStack failure. We decided it should be investigated futher so we saved the data from clean up and filed a ticket on the problem.
  • installed a new ipp at MPIA in Heidelberg.

Chris Waters

  • Out one day due to illness.
  • Stacking: finished improvements to rejection in stacking. A small test shows that artifact blobs around bright stars are cleared up. Continued testing to fix remaining bugs.
  • Asinh scaling: finished and committed to trunk.
  • Diskspace: queued LAP warps on the distribution server for cleanup. Optimized neb-shuffle code to speed up transfer of data to ATRC. Culled laser/calscreen data down to only retain a single copy.
  • Worked on math to convert EXT_NSIGMA into probabilities of CR/star/galaxy for PSPS.