Monday Morning Meeting 25 May 2017

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

Attendance

  • Socorro: Juergen, Jeff, Lindsey, Kumar, Tak, Martin, George, Urvashi, Pam
  • CV: Dave, Darrell, Anand, Morgan, Akeem, Andy, Bob
  • Garching: Dirk,
  • SCO:

News / Meetings / Visitors

  • Jeff in CV next week, ANASAC and UC
  • NRAO Holliday Monday
  • ESO Developers Visit June 5-9
  • CARTA Meeting June 19-23
  • SKA Meeting -- Meeting between DMS mgmt and SKA to create an MOU for advancing the data model for CASA -- make CASA scale better for parallelization. Agreed to work on MSv3. Our current HPC effort scales to hundreds of cores. SKA needs a model that scales to thousands of cores. Timescale is a year. This will be run as a subproject of casa. However, SKA will be doing the work and mostly managing the project.
  • VLASS CDR was this week. Still technical requirements that are in flux. Not many action items likely to come out of the CDR for CASA. There are 2 blockers: additional DA support needed; discrepancy in polarization calibration must be understood.

Build, Release

  • 5.0 status
    • User testing of tclean on going.
    • Problem with frequency axis.
    • Mask / tclean issues
    • CAS-10110
    • Documentation display issues that need to be fixed.
    • Lindsey's doc error messages: we depend on web browser modeule from python; it depends on stuff that Linux/OSX provides for opening browser. It might be good to catch the error and generate a human-readable error message.
    • Final documentation review on week of June 5th.

Verification Testing

  • Moved one of the test Macs to a build Mac
    • Needed to keep up with Bamboo builds
    • OSX 10.12
  • CASA guides are not working, so for now they have been taken out of the test suite
    • Certain machines only see data-local, but CASA guide data is in data-shared
    • Will be put back in after the 5.0 release
  • Both master and prerelease are looking at the same test directory
    • New tests in master will not work in prerelease
    • New tests will be put in after the 5.0 release
    • There will be some discussion within the build and test teams, and decisions made on how to manage this issue without impacting the CI system

Validation Testing

  • Testing of VI/VB@, tclean, calibration

Architecture

HPC

Development

  • CASA 5.1 Planning: VLASS, pipeline, asynch I/O stuff.
    • There are 256 miscellaneous tickets.
    • If you don't know what you should be doing for 5.1, talk with Jeff, Morgan, or Anand.
    • Asynch I/O: Jeff wants to do CDR in 5.1 cycle; will not be turned on by default, but would like to have it ready for testing just after release.
    • 5.1P1 pipeline branch has been created.

Pipeline

AOB

  • Eduardo Miranda's Report wrt Casa Group Lead
    • Darrell: Miranda seemed to recommend giving more power to project scientist (and lower level staff) to free lead to do long term planning.
    • Morgan is looking for a lead who is somewhere between a dictator and a coordinator.
  • CASA Lead search committee. Please feel free to share your ideas with members.
    • Technical Committee Members: Urvashi, Lindsey, Darrell, Rafael;
    • Stakeholders Committee: Kumar is a part.
  • Interim Coordination Team
    • Members:
      • Darrell - build / system
      • Urvashi - science
      • Jen - testing
      • Juergen - project scientist
      • Lindsey - pipeline
      • Anand -
      • Morgan - interim lead
      • Sandra - ESO
      • Kana - NAOJ
      • Shin Fey Lee - ASIAA
    • Andy was considered for this group, but mostly he will not need to attend.
    • This type of management team was tried in the Rob Selina era. It worked.

Developer Reports

Monday Meeting
  • Sanjay Bhatnagar
  • Sandra Castro
  • Lindsey Davis
    • Release note preparation
    • Updates to pipeline manifest and aqua report generator
  • Pam Ford
  • Enrique Garcia
  • Bob Garwood
    • ALMA code move related work - disentangling code generation steps from makefiles, pushing forward on syncing ALMA and CASA versions before the move.
    • Some GBT-related work.
  • Kumar Golap
  • Jeff Kern
    • Preparation for UC
    • VLASS CDR
  • David Mehringer
  • George Moellenbrock
    • VLA tour last Thursday for Socorro 4th grade. (Thanks also to Kumar for helping!)
    • Re-examined CAS-6130 and CAS-8527: there are still some problems with cal library relating to lack of availability of calibration for all spws known to an MS. Best to fix this in 5.1.
    • 5.1 JIRA planning
  • Dirk Petry
  • Martin Pokorny
  • Federico M Pouzols
    • Thursday 25 is a holiday in Germany
    • CAS-5808: "rflag's use of timedev and timedevscale parameters", ready to verify/test.
    • CAS-9089: automatic generation of antenna position corrections from web service / ALMA TMC DB. Service time out issue just fixed after our ICT leads promoted ticket to "blocker": http://ictjira.alma.cl/browse/ICT-9872. Hoping to get this ready for testing next week.
  • Urvashi Rao
  • Darrell Schiebel
  • Ville Suoranta
  • Takahiro Tsutsumi
    • Checked in a temporary fix for tclean automask issue on nfs mounted disk for 5.0
    • Looked into on tclean 's cube image setup difference between 4.7.2 and 5.0 (CAS-10194)
Friday NAOJ Meeting
  • Kanako Sugimoto
    • Fully migrated to Eclipse + EGit development environment thanks to instruction wiki by Takeshi.
    • Appropriate UNIT handling in singledishfiller ( CAS-10043): import part is done in Nobeyama and scantable filler. CORRECTED_DATA column generated by calibrator tool does not respect optional UNIT keyword while mstransform does. Need to decide what to do.
  • Wataru Kawasaki
  • Masaya Kuniyoshi
  • Takeshi Nakazato
  • Renaud Miel
    • CAS-8082 plotms header: All requested header items implemented. Screenshot

-- KanaSugimoto - 2017-05-25
Topic revision: r19 - 2017-06-01, JaredCrossley
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