Casa Parallelization meeting minutes

Monday November 15th Room 317, 3:30pm MDT

Attendees: Kumar, Jim, Brian, Jeff, Greg, James

Targets/tasks for November 15 - November 21

• Begin code development of async I/O Jim
• Complete remaining timing scripts Greg
• Generate procedures and first timings for test cases James
• Time domain partitioning with multiple MSes per imaging process Kumar

Discussion/Agenda

• Identify somebody to test scan selection in filler Jeff/Brian
  • Done, seems to work. No immediate need for task, debate on whether it should get into active only.
• Create async I/0 design doc Jim
  • Discussion of threading selection Boost vs Posix. Posix for now
  • To the extent that any one read the design doc it looked good.
  • Opt in approach, begin code development
• Complete timing scripts Greg
  • Small spectral line still needs some cleaning.
  • Small continuum is done, testing in progress
  • Large ones to follow
• Generate procedures and first timings for test cases James
  • Split is very brutal in casapy. Test against casapy-test
  • Starting to look at. (split, imaging, applycal, clip in that order seemed to be most expensive tasks)
  • 3s averaging in spectral case 10s averaging in cont
  • Some discussion on whether averaging/split etc is really necessary in the future. Consensus is leave it in for now, it represents current reality and we should at least be aware of performance issues.
• Time domain partitioning with multiple MSes per imaging process Kumar
  • In +Testing
• 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