Monday Morning Meeting 01/22/18

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

Attendance

• Socorro:
• CV:
• Garching:
• SCO:

News / Meetings / Visitors

• Klara Shabun start date 1/23
• CASA F2F agenda ideas/discussion - https://safe.nrao.edu/wiki/bin/view/Software/CASA/CASAF2F_2018Feb
• Build/Test Process Review
  • CASA review completed - feedback used to plan 5.4 process updates
  • Pipeline review TBD
  • Feedback page comments resolution still owed
• Development Process updates (beginning with 5.4)
  • CASR project for requirements management
  • CAS project for development only
    • Planning to reduce number of ticket types
    • Planning to reduce number of stages in ticket workflow
  • Confluence Cleanup
    • High level work tracking of release cycles (next evolution from 5.3 page)
    • Designs and working data for longer term development items (looking at Creately plugin)
• Confluence Calendar Cleanup

Development

• CASA 5.3 Progress Tracking: https://open-confluence.nrao.edu/pages/viewpage.action?pageId=7046483
• CASA 5.3 Ticket Summary: https://open-jira.nrao.edu/secure/Dashboard.jspa?selectPageId=11005

Pipeline

• Pipeline F2F likely to be May 7~11
• CASA 5.1 / Pipeline
  • Complete
• CASA 5.2 / Pipeline / HPC
  • Complete
• CASA 5.3 / Pipeline
  • CASA / Pipeline callibrary integration
    • standard ALMA interferometry use case works
    • issue with undefined / missing calibrations and calflagstrict
  • use of atmospheric plot overlays in plotms waiting in CASA plotms validations
  • plotcal to plotms migration, prototyping in progress
  • support for multi-vis ephemeris images waiting on CASA
• Pipeline
  • implementing support for multi-vis matching of spws by name rather than id in imaging code
  • support for antenna selection in imaging code, for dealing with mixed array observations
  • single epoch imaging for VLASS

HPC

• CASA 5.2 (prerelease-5.2.0-5)
  • General performance issues fixed - CAS-10794.
  • No issues found with several nodes / many cores.
  • Looks ready. There are open issues for the future:
    • CAS-10857 - "Differences in pipeline runs between serial and parallel (HPC) testing - hif_findcont and hif_makeimages" (tclean weights and findcont sensitivity to parallel/serial differences)
    • Differences in warnings and error messages between serial and parallel runs.
• 5.3 - Development, bugfixes
  • Discussed the possible solution for the VLA problems with uvcontsub. CAS-10697: uvcontsub does not work in parallel mode with 'combine=spw' - in progress
  • Created CAS-11076: mstransform needs a way to create a normal MS from an MMS

Users

• New benchmarking-plots CASA lustre/naasc/ serial vs parallel attached
• Revised CASA website online
• European Week of Astronomy and Space Science 3-6 April: slot for talk available (D. Petry), but would need volunteer.

Build, Release

Verification Testing

Validation Testing

Architecture

• Warnings Summary

AOB

Developer Reports

Monday Meeting

• Sanjay Bhatnagar
• Sandra Castro
  • 5.3 Development
    • CAS-10568: Plone content: Global Task List - flagcmd - Ready to Verify
    • CAS-10310: doc strings need to be updated to reflect "MeasurementSet" - Ready to validate
    • CAS-10383: flagdata should fail if datacolumn=RESIDUAL and MODEL or virtual MODEL columns don't exist - ready to validate
    • CAS-10697: uvcontsub does not work in parallel mode with 'combine=spw' - in progress
    • CAS-4888: flagdata functional tests should test with MSs that contain virtual model columns - in progress
• Lindsey Davis
  • Resumed work on ALMA pipeline infrastructure code
    • Installed current software, and reactivated desktop test environment
    • Installed local git repository, investigating current branching "standards"
  • Verification testing of CASA callibrary changes
  • Pipeline input parameter refactoring and testing (vdp parameters), working for ALMA, some VLA issues
• Bjorn Emonts
• Pam Ford
  • continued 5.3 work for plotting CalTables (BPOLY, GSPLINE) in plotms
• Enrique Garcia
• Bob Garwood
• Kumar Golap
• David Mehringer
• George Moellenbrock
• Dirk Petry
• Martin Pokorny
• Federico M Pouzols
  • Miscellaneous mstransform/cvel tickets, WIP: CAS-10446, CAS-10584.
  • CAS-11059: clarification of 'width' parameter in mstransform/cvel2.
  • PRACE course at LRZ Leibniz Supercomputing Center: Introduction to hybrid programming in HPC, MPI/OpenMP
  • Going back to changes to flagging/rflag interface and documentation: CAS-5808, CAS-5809, CAS-10036.
• Urvashi Rao
• Darrell Schiebel
  • CASA++
    • CASAtools
      • builds with Python's setup.py & GNU autoconf
      • builds with Python 3.* and 2.7
    • CASAviewer (current "casaviewer" built and distributed stand-alone)
      • builds with qmake & minimal config
      • builds with Qt4 or Qt5
      • working on packaging with AppImage (linux) & regular OSX App
• Ville Suoranta
• Takahiro Tsutsumi

Friday NAOJ Meeting

• Kanako Sugimoto
  • sideband separation task, sdsidebandsplit ( CAS-8091): Done and under validation
  • add reindex option to sdbaseline and sdsmooth ( CAS-10789): Done and under validation
  • duplication of FITS keyword in images generated by sdimaging (and clean) ( CAS-8248): Done and under validation
  • update plotbandpass to support Nobeyama ( CAS-11031): Transferred updates by Todd in analysisUtil to the CASA task. Branch merged.
  • Plone document of sdsidebandsplit ( CAS-11041): started the work
• Wataru Kawasaki
• Takeshi Nakazato
  • management
    • start discussion about 5.4 requirement decision
    • updated documentation for sdms tool based on reviewer's comments ( CAS-9706, thanks Wataru)
    • resolved CAS-7320feed selection
      • validation was basically done using artificial data
      • there is no hope to get real data useful for testing
  • CAS-10927 bug fix for undefined spectral reference frame of image produced by tsdimaging: Under Validation
  • CAS-9956 performance degrade of sdcal (calmode='otf'): RESOLVED
  • CAS-11056 performance degrade of sdcal (calmode='apply'): RESOLVED
• Renaud Miel
  • CAS-8087plotms pointings
    • • decision was made with Takeshi that implementation should be done by implementing a TVI. TVI implementation ongoing.