Casa Parallelization meeting minutes

Monday November 22nd Room 317, 3:30pm MDT

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

Targets/tasks for November 22 - November 29

• Verify existing small data scripts Sanjay
• Continue 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

• Begin code development of async I/O Jim
  • Initial unit tests underway
  • Functional version should test parallel imaging task first
• Complete remaining timing scripts Greg
  • Small spectral status (needed further cleaning)
  • Large continuum status
  • Large spectral status
• Generate procedures and first timings for test cases James
  • Split dominating continuum case
    • Continued debate about usefulness of split and concat in the first place
    • Discussion of whether split is appropriate issue for this group.
    • Agreed it should remain in scripts but be bypassed for general testing.
  • Tile size effects in spectral line, agreed to redo spectral case to 1MB tile.
    • Kumar to create 1MB tile for spectral line test
  • Some initial results are done, difficult to work past noise of split in continuum case
  • Need other data sets
  • Sanjay to look at scripts and verify them
• Time domain partitioning with multiple MSes per imaging process Kumar
  • Testing in progress, timing tests showed multi-ms slower to single ms.
• 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