Overview of EVLA Data Post-Processing: Goals and Challenges from the December 2004 Meeting
This document summarizes the key agendas discussed during the EVLA Advisory Committee Meeting held on December 14-15, 2004. It covers project goals such as fulfilling requirements for post-processing, meeting delivery schedules, ensuring robustness and usability, and maximizing resource efficiency while integrating development efforts with ALMA. It also highlights the challenges faced in wide-band imaging, data rates, and user interface development. The roles and contributions of various algorithm and user groups, including their focus on improving functionality and supporting current VLA users, are also outlined.
Overview of EVLA Data Post-Processing: Goals and Challenges from the December 2004 Meeting
E N D
Presentation Transcript
EVLA Data Post-processing Overview Steven T. Myers AIPS++ Project Scientist EVLA Advisory Committee Meeting December 14-15, 2004
Goals • Fulfill project requirements • post-processing requirements (Butler, Myers) • Meet delivery schedule • management plan (McMullin, Greisen) • Functionality, Robustness, Usability • integrated development & testing plan • Efficient use of resources • maximize co-development with ALMA EVLA Advisory Committee Meeting December 14-15, 2004
AIPS • extensively developed & well-tested • will be used in commissioning • avoid commissioning both telescope & software! • plan: continued support + targeted development • support current VLA users • develop new functionality (e.g. auto-flagging, RFI) • support as much of EVLA processing as is practical EVLA Advisory Committee Meeting December 14-15, 2004
aips++/casa • primary support for full EVLA processing • maximize co-development with ALMA • most core functions are in common • share development team & management • merge EVLA and ALMA development plans • EVLA focus on “Delta” problems • EVLA testing during development cycles • ALMA tests already use VLA data! EVLA Advisory Committee Meeting December 14-15, 2004
Usability & Interface • Interface • current focus on CLI and scripting for functionality • necessary special purpose “GUI”s (e.g. viewer) only • General GUI development • defer to new Framework development(s) • scheduled for 2007 • Usability is key for EVLA user acceptance • rely on scientist input (e.g. NAUG) EVLA Advisory Committee Meeting December 14-15, 2004
Problems • Difficult EVLA problems • wide-band wide-field full polarization imaging • high dynamic range high-fidelity imaging • also: RFI, pipelines • Data rates • 2008 spec 25 MB/s max (cf. VLA 0.1 MB/s) • WIDAR can produce much higher rates! • data volumes also (TB datasets) EVLA Advisory Committee Meeting December 14-15, 2004
Algorithms • Algorithms group • led by Bhatnagar & Myers (formerly Cornwell) • includes AIPS & aips++/casa developers, students • NRAO-wide staff participation (e.g. NAWG) • Progress • w-projection, pointing corrections (TC, SB, KG); • autoflagging, RFI (EG, BC); pol. beam (TC,SG,KG); • RFI (TC,RP,SB,KG); MS-clean (TC,SB,KG,EG) EVLA Advisory Committee Meeting December 14-15, 2004
Risks & Issues • difficult problems (e.g. wide-band MFS) • schedule (tight given current resources) • data rates & size cluster computing? • user interface (“green hole”) • aips++ support for VLA? or EVLA only in 2007? • pipelines: afford development & ops support? • new frameworks (ACS? general? developer?) EVLA Advisory Committee Meeting December 14-15, 2004
Appendix: Groups • NAUG = NRAO Aips++ User Group • aips++/casa programmers and scientific staff • currently ~1 FTE of testers (AOC & CV) • pre-testing for ALMA and EVLA • use cases and development guidance • NAWG = NRAO Algorithms Working Group • started Nov 2005, scientific staff NRAO wide • regular forum to discuss algorithm issues EVLA Advisory Committee Meeting December 14-15, 2004
