Monday Morning Meeting 12/4/17

  • DIAL-IN NUMBERS & PASSCODES:
  • IP: 192.33.117.12##8110
  • Phone: (434) 817-6524

Attendance

  • Socorro:
  • CV:
  • Garching:
  • SCO:

News / Meetings / Visitors

  • CASA face to Face - Feb 26
  • Build/Test Process Review - CASA Developers Meeting 12/6
  • CASA 5.3 Status and Tracking

Development

  • CASA 5.3 --- Calibration (GM)
    • Polarization (CAS-9105)
      • Basis-agnostic cross-hand phase (CAS-9108,9112,9114)
      • Position-angle solver for linear basis (CAS-9113)
      • Streamlined QU solvers for both bases (CAS-9106,9107)
      • Discussions re non-linear D-solver for Stokes V and ALMA "short scheme"
    • Cal Library
      • Alignment with ALMA PL usage (w/ Lindsey, CAS-10723)
      • Discussions ongoing with Juergen/Lindsey on improvements
    • VLBI (mostly at JIVE)
  • CASA 5.2
    • CAS-10849 : Restoring beam : Looks ok from CASA's side.
    • Refconcatimage slowdown/speedup : Update from KG ?

Pipeline

  • CASA 5.1 / Pipeline
    • Issues with the current restoring beam heuristics which are the result of a tradeoff between algorithm purity and efficiency
      • Problem: heuristics rely on an unflagged central channel for cube images, and reuse in the continuum case
  • CASA 5.2 / Pipeline
    • CAS-10849: problem implementing the suggested heuristics possibly due to restart option used by tclean calls
      • (UR : Does the new verification test run and pass in the tarball you are using ? Update : D.M. confirmed that he was using the wrong tarball)
  • CASA 5.3. / Pipeline
    • VLA pipelines:
      • QA score implementation, AQUA report support, easier testing
      • Ongoing plotms issues
    • Discussions on CASA / pipeline callibrary integration issues resumed
    • Plotcal -> plotms migration issues

HPC

Users

  • Updated CASA website: https://casa.nrao.edu/index_testcopy.shtml
  • Stats viewer issues: 17% of all NRAO helpdesk tickets related to visualization, mostly viewer (% for ALMA helpdesk substantially lower). Of these 17% critical/bugs, 44% major, 28% minor, 11% user 'error'. Bugs/critical: HD10827: plotms crashes under certain condition --> fixed in 5.1; HD11209: plotting rotation angles wrong; HD11237 Table array conformance error for data set with different size spw, even if only 1 spw is plotted

Build, Release

Verification Testing

  • Emergency update on the two OSX 10.13 machines due to a gaping security flaw
    • Since NRAO does not officially support 10.13, it's up to us to manage these machines
  • Replied to Enrique's entry in to the build and test process feedback page: https://open-confluence.nrao.edu/pages/viewpage.action?pageId=7045225
  • Test team is focused on finishing up ALMA phase B testing this week, so limited availability

Validation Testing

  • Current validation tally
    • 122 Under Validation, 87 Ready to Validate.
    • Nov 17 - 23: 7 tickets assigned for validation. 3 tickets resolved.
    • Nov 24 - 30: 7 tickets assigned for validation. 6 tickets resolved.
  • JDM working largely on WVR project through Dec 15

Architecture

AOB

Developer Reports

Monday Meeting
  • Sanjay Bhatnagar
  • Sandra Castro
  • Lindsey Davis
    • Added support for "import as imaging measurement" set to the pipeline importdata tasks with default=False
    • Added check for empty pipeline callibrary to pipeline applycal tasks
    • Changed interpolation for calmode='a' solint='inf' tables from linear,linear to nearest,linear, pipeline callibrary handles same table different apply rules transparently
    • Callibrary discussions with George and Juergen
    • ICT Archive / Pipeline tickets refactored into separate archive and pipeline error handling tickets
  • Bjorn Emonts
  • Pam Ford
    • CAS-10900: Completed CTSummary class for plotms summaries for cal tables, ready for testing
    • CAS-10774: Failed to reproduce pipeline plotms failures which appear to be plotms crashes (plots succeeded for me). Contacted others to request enabling core files for user vlapipe. Notified Darrell that crash report poster is not working on el6.
    • Pull requests merged for CAS-10818 (ms.selectinit with invalid DDID) and CAS-10848 (ms.getdata with capitalized item) bug fixes
  • Enrique Garcia
    • Working on CAS-10458: Vi/VB2 vb->visCubeCorrected() segfaults if no column. Improving unit tests on VI/VB2 while at that.
    • Fixes on casacore related to MsFactory MS simulator.
  • Bob Garwood
    • CAS-6832 importasdm should be able to import data up until the first corrupt scan. This was a problem for some cycle 2 data (ICT-331). The appropriate fix to asdm2MS is being negotiated.
    • CAS-10949 Change the behavior of asdm2MS when ASDMBinary file(s) are missing. Ongoing discussions.
    • CAS-10386 Add new shared ALMS code to support 2 new ASDM tables in asdm2MS. Testing with new Pulsar table from the VLA. Appears ready for validation.
    • CAS-10644 Filler needs to detect duplicate (sets of) rows and retain only one copy. Working for Pointing, back to working on this problem for the main table for WVR data (duplicate DATA values).
  • Kumar Golap
  • David Mehringer
  • George Moellenbrock
  • Dirk Petry
  • Martin Pokorny
    • Progress on asynchronous VI2: basically working, now addressing performance issues
  • Federico M Pouzols
    • Profiling and trying to understand CAS-10795, casacore storage managers, etc.
  • Urvashi Rao
  • Darrell Schiebel
    • configuration of grpc
    • experiments with casacore startup initialization
  • Ville Suoranta
  • Takahiro Tsutsumi
Friday NAOJ Meeting
  • Kanako Sugimoto
  • Wataru Kawasaki
  • Takeshi Nakazato
  • Renaud Miel
CONFLICT original 12:

-- RyanRaba - 2017-12-01
CONFLICT version 13:

CONFLICT original 9:
CONFLICT version new:

-- RyanRaba - 2017-12-01
CONFLICT end

CONFLICT version new:

-- RyanRaba - 2017-12-01
CONFLICT end
Topic revision: r16 - 2017-12-04, MartinPokorny
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback