Casa Parallelization meeting minutes

Monday March 18th Room 317, 3:30pm MDT

Attendees:

Targets/task

  • Remaining partition non-compliant tasks
    • Status update on tasks that still need work to be partition compliant
    • MS transform in good shape (replaces split, cvel, hanning-smoothing etc)
      • Does not do uvconsub
      • Potentially testable in May
    • Move virtual model column to source table * Filler
    • Status and future plans discussion
    • Looking at both parallel and new filler approiach and storage manager
  • Parallelization of MS-MFS major cycle
    • Urvashi is working on as to do for 4.1 with more complete solution after 4.1 * In progress
    • Will include new FT machines, mosaicing, A-projection added to MS-MFS
  • Other
    • Parallelize Cube imaging scatter gather (low level target for 4.1)
      • Switch to C++ concat images rather than python
      • Some parallelization
    • Expand threaded gridder case to >4 threads (not a 4.1 deliverable)
      • Sanjay looking at LOFAR approach
  • Resource requirement heuristics by task (imaging first) targetted for 4.1
    • Discussion of how to proceed with making task resource (memory, number of cores etc) aware.
    • Large channel count cube imaging
    • Frist draft of heuristics document
  • Issue with temp lattice calculation

Discussion

  • Temp lattice
  • Clean Parallelization heuristics

Deferred Items

-- JamesRobnett - 2012-12-17
Topic revision: r1 - 2013-03-27, JeffKern
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