Monday Morning Meeting 27 February 2017

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

Attendance

  • Socorro: Jeff, Martin, George, Lindsey, Tak, Pam, Kumar
  • CV: Dave, Darrell, Ville, Bob, Andy, Akeem, Jen, Anand
  • Garching: Fredrico,
  • SCO:

News / Meetings / Visitors

  • CASA Newsletter -- time to begin working on this; release will be in April.
  • No Meeting March 13
  • Ville part 2 tomorrow

Build, Release

  • CASA-4.7.2
    • bug found in 4.7.1 that is fairly severe. We will be making a 4.7.2 release. First build was made over the weekend.
    • Will also include upgraded antenna configurations for ALMA
  • CASA-5.0 -- We've given ourselves schedule slack because we brought in some things late: tclean and calibration
  • Discussion on updating matplotlib: plotants and plotcal have to be dealt with.

Verification Testing

  • CASA guides: CISupdate to Mediawiki at the end of the month will break the script extractor
    • Implications?
    • Proposed workaround is to create static scripts that will be stored in repository. These will be used for testing.
    • Jeff is not in favor of the upgrade. Jeff will discuss with CIS directly.
  • Discussion on new build and test hardware on Friday
    • Linux Vms (2 RH6, 2 RH7) - one RH7 VM dedicated to docker testing
    • 4 additional Mac Minis - one will be dedicated for next OSX release testing/building
    • Repurposed cvfiler to "host" data repo - testing will be done to make sure we are gaining performance
    • Docker testing will be compared with current VM build/test model to make sure results are consistent, and hopefully test times and resource overhead in general will be reduced

Validation Testing

  • current ticket status 97 RTT, 146 UT. the majority of both lists are plone related.
  • a ton of assignments have gone out and are starting to turn around with feedback, all due nominally on 3/1, so be aware that you'll likely be getting suggestions to implement or questions to answer on your documentation.
  • once the plone madness abates a little, I'll be working on tackling everything else.
  • We will not be able to do a full review of the documentation; Juergen will take a last look over the whole of the documentation, but this may not be deep.

Architecture

HPC

Development

  • Housekeeping
    • Jeff does not see a reasonable way to coordinate this across the whole team. However, we should schedule a housekeeping focus on specific topics. VI/VB2 will be the first topic:
      • There are rough edges that need to be dealt with.
      • There will be a talk by George in late March; following this by a month or less, we will have a meeting regarding housekeeping.
      • The plan is for Martin to maintain the VI/VB2 code. Right now, Jim is still supporting this.

Pipeline

  • CASA 4.7.1 tclean model column issue
  • CASA 5.0 testing
    • added mycb.setvi(old=True, quiet=False) fix to pipeline uvcontfit task, seems to be working
    • vi/vb2 tclean problems for spw(s) where channel frequency decreases with channel number
    • HPC problems
  • ALMA pipeline workflow discussions
  • Development
    • EVLA polarization heuristics working correctly, lots of small EVLA task changes
    • image list editing task testing

AOB

Developer Reports

Monday Meeting
  • Sanjay Bhatnagar
  • Sandra Castro
  • Lindsey Davis
    • implemented ALMA representative source identification heuristics
    • CASA 5.0 pipeline testing
      • uv continuum fitting fix, scipy / numpy changes issues in find continuum code, tclean problems
    • ALMA pipeline workflow discussions
  • Pam Ford
    • Investigating averaging questions in CAS-9682 - scalar averaging and time-averaged derived values (e.g. phase)
  • Enrique Garcia
  • Bob Garwood
    • Deep dive into the ASDM and code generation process and Michel's first pass at separating that from the rest of the ACS code tree.
    • Continued effort to understand bdflags2MS
    • Started looking at set of CASA tickets assigned to Michel. Michel has promised me a summary of what he thinks the status of each one is.
  • Kumar Golap
    • Mainly vi/vb2 fixes.
    • Waiting on Jim to fix stuff.
  • Justo Gonzalez
  • Jim Jacobs
  • Jeff Kern
    • Requirements Derivation for StatWt
    • Usual Meetings
  • David Mehringer
    • Working with Jim to get a VI/VB2/TVI document that developers with no experience in this framework can understand.
    • Working with Jeff to distill necessary requirements for statwt.
    • Got lots of JIRA emails for various documentation tickets changing hands.
    • Reassigned many statwt tickets to Anand while necessary requirements and design docs are written. I believe Jeff has since closed some of those which are not relevant to the rewrite.
    • (CAS-9195) RMS images. Replaced three high level calls with numerous low level calls to improve performance by of order a factor of 50. Implemented and committed task interface.
    • (CAS-8963) investigate imstat performance issues. Asked tester to summarize most egregious cases of slowdowns from 4.6 to 5.0.
    • (CAS-5740) CASA needs equivalent to BLSUM/ISPEC for integrated line flux density. Fixed reported bug re: the stretch parameter.
    • (CAS-9130) impv improvement. Asked submitter for clarification of request.
    • (CAS-9490) ia.rotatebeam() should at least have the option to create a new image. Asked Juergen if it is OK with stakeholders to change default behavior as requested in ticket.
  • George Moellenbrock
    • Working on vi/vb2 calibration stuff. Sorted out number of minor things; not performing as well as old version.
  • Dirk Petry
  • Martin Pokorny
    • Reviewed vi/vb2 stuff. Now looking at where the asynchronous code was left.
    • We will develop the requirements and do a light weight CDR. We will do this as one of the Tuesday meetings.
  • Federico M Pouzols
    • Investigating and working on mstransform and selection:
  • Urvashi Rao
    • Documentation; imaging documentation 90% complete. Will not be 100% perfect for quite a while.
    • Jen will be assigning Urvashi documents very soon.
  • Darrell Schiebel
    • looked at crash reporter
    • resolved remaining documentation CLI issues
  • Ville Suoranta
    • Carta help/discussions
    • Worked on Crash Reporter
    • Moved Macs to the server room
    • Bitbucket/Bamboo transition meeting
    • Bitbucket workflow tests
    • Build management
    • ic2233?
  • Takahiro Tsutsumi
    • Looked at some bug tickets for 5.0
    • Minor plone doc edit
Friday NAOJ Meeting
  • Kanako Sugimoto
    • bug fixes to importnro ( CAS-9485): I believe all fixed
    • discussion of 1-base refChan in ATM library: All the other part of code is 0-base. Possibly the left over of Fortran era. I can fix it in tool level but it does not help for codes that directly use ATM library, e.g., AtmCorruptor. Waiting for Juan's feed back but so far no response.
  • Wataru Kawasaki
  • Masaya Kuniyoshi
  • Takeshi Nakazato
    • bug report and fix of casacore::Convolver problem (GitHub casacore issue #572)
    • two engineering tickets for NRO development: CAS-9708, CAS-9763
    • potential bug fix on importasap and importnro: CAS-9775
    • fix for the error reported by tester: CAS-7904 (polaverage)
  • Renaud Miel
    • CAS-8082 PlotMs Header: implemented/tested passing of header items parameters from casa/python to casaplotms through Dbus
    • CAS-9344 importasap Plone doc: reviewed and commented on ticket

-- KanaSugimoto - 2017-02-22-- KanaSugimoto - 2017-02-22
Topic revision: r17 - 2017-03-06, 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