Casa Parallelization meeting minutes

Monday January 24th Room 317, 3:30pm MDT

Attendees: Jim, James, Kumar, Bryan, Sanjay

Targets/tasks for January 24th - January 31st

• Continue code development of async I/O Jim
• Time domain partitioning with multiple MSes per imaging process Kumar
• Generate large cont script with async I/O Kumar
• Roll split improvements into Casatest

Discussion/Agenda

• Continue code development of async I/O Jim
  • working toward getting imager to make use of it, difficulties in integration
  • Test that it only uses async if explicitly told to
  • Need to check in code and test that casarc script works to trigger sync vs async
  • Potential problem with pulling copy within recursive iteration through visiter
• Complete remaining timing scripts Gregg
  • Small spectral status (needed further cleaning)
    • does not include continuum subtraction as a separate timed step using uvconsub2.
  • Large continuum essentially done
  • Large spectral
• Generate procedures and first timings for test cases James
  • Split seems much better, need to prove correctness
  • Tile size substantially affects performance, need to characterize behavior
  • Still need other data sets
• Time domain partitioning with multiple MSes per imaging process Kumar
  • Cube case regression status
  • Trying to identify 10% small reads, possibly metadata column (either main or sub table)
• Rob Reid in NM
  • Significant progress in understanding nature of split I/O problem, Visibilityiterator vs visiterator
• Michel Cailat in NM
  • Progress in understanding nature or flagging I/O problem. Test code provided that read BDF without mapping into memory.
  • Much faster, much lower memory requirement
• Cluster scheduler scripting with CV James on hold for scripts
  • Discuss with Jim semantics for cluster/CASA interaction
• Cluster resource definitions James/Jim on hold