ALMA SSR Committee Phone Meeting 2003-09-10

Connection Details

• Call date: 2003-09-10 (Wednesday)
• Call time: 15:00 UT
• Duration: 1 hr
• USA Toll Free Number: 888-795-6990
• Toll Number: +1-484-630-4370
• Passcode: 99640
• Leader: Robert Lucas

Attendance

Minutes

1. Subsystem Scientist Activities (all)
   
   
   • Jeff Mangum -- Control
     Continue with ATF activities serving as proving ground for ALMA control system.
     
   • John Richer -- Telescope Calibration
     
   • Peter Schilke -- Data Reduction User Interface
     
   • Steve Scott -- Correlator Participated in the Correlator Subsystem CDR1 in July. Post CDR work with Jim Pisano on modifications to the Design and ICD documents.
     
   • Debra Shepherd -- Off-Line Analysis
     (for the last two months)
     • Helped to write the following Use Cases in collaboration with subsystem leads and SSR members:
       • Control.UC.automatic.html Automatic Operations for control.
       • ObsTool.UC.MultiFieldMosaicSetup.html Multi-Field Mosaic setup.
       • Offline.UC.TotPower.html Reduce & Image Auto-Correlation Data
       • Sched.UC.dynamic.html Automatic scheduling operations
       • Sched.UC.interactive.html Manual scheduling operations
     • Still developing Use Cases with Pat Palmer:
       • ObsTool.UC.SolarSystemObsSetup.html Comet observations setup
       • Offline.UC.SolarSystem.html Comet data reduction
     • Received Use Cases from other subsystems (Executive, ObsPrep, Pipeline, Telcal) and put these on the web. Also created a page to hold Test Plans: http://www.aoc.nrao.edu/~dshepher/alma/testplans Reviewed ObsPrep and Pipeline test plan drafts. Updated the Offline testing time line.
     • Wrote a summary of the SSR guidelines and expectations for the AIPS++ benchmark effort. This is available at: http://www.aoc.nrao.edu/~dshepher/alma. Continued to work with the AIPS++ group to create automatic benchmark results. Note: the gravitational lens benchmark dataset has been expanded via simulation to have longer processing times so comparisons will not be dominated by set-up times. The SSR & community-supplied scripts are being compared with each other by the AIPS++ team to ensure that each processing step is identical or differences are documented. Comparison numbers between AIPS++ and GILDAS using the Pseudo-GGTau dataset are now available on the web at: http://shiraz.drao.nrc.ca:8080/AlmaDRPBenchmarks/
     • Attended two AIPS++ meetings/week to track development plans along with the monthly NRAO software testing group meetings. Worked with Joe McMullin to coordinate AIPS++ development priorities with the software testing requirements.
     • Created a list of all offline 'U' requirements and sent these to the SSR so we know what needs to be worked on. Most of the uncertainties have to do with a lack of definition of standard ALMA modes (which the Science IPT is working on). Other 'U' requirements are related to benchmarking and these will be re-evaluated as soon as the benchmark effort is established.
     • Reduced and imaged NH3 line data in AIPS++. Started exercising the types of analysis functions that will be needed for the Jan04 Offline test (so I can write a reasonable cookbook for the testers). Submitted one enhancement request relating to region defaults on channels during interactive clean. Found 1 minor bug (zoom during interactive clean) and 1 limitation with channel selection while creating moment maps. Talked to Joe M. about these. Submitted bug report/enhancement requests. Also submitted an enhancement request to make imager defaults more useful.
     • Finished modifying the AIPS++ cookbook for VLA data reduction. Deleted all GUI instructions & changed/added some figures. Gave the document to George Moellenbrock & Steve Myers who will now be responsible for document upgrade/modification. This cookbook will be used by the NRAO software testing group who will do pre-testing for the ALMA Jan04 user test.
     • Attended the ASAC meeting in Hamilton, Ontario, Canada on 5 and 6 Sep03 with Brian Glendenning. Presented the ALMA software Test Plan definition and status along with a summary of recent AIPS++ improvements (including recent enhancements to the calibrator and imaging tools as well as the automated benchmark effort).
     
     
   • Leonardo Testi -- Observing Program Preparation
     With the help of Debra, I prepared a number of use cases for testing purposes. A first draft of the test plan was discussed with Alan Bridger and Debra Shepherd, I plan to include their comment and finalise the first year plan within next week. The focus of the first test session will be discussed with the subsystem team at our face-to-face meeting in Florence next week.
     Had a couple of telecons with the team in July and August, we discussed the CDR1 result and the progress toward R1 and some technical issue, no major problems.
     An issue that keeps jumping in our discussions and reviews (PDR, CDR1) is the how much PhI and PhII should be integrated together. I believe that we have always been convinced that the user should be able to dig into the Phase II details even when preparing the Phase I material, if willing to do so. This has some consequence on some technical aspects of the design (that is why it keeps resurfacing), something related to creating same (and conflicting) entities at Phase I and Phase II. I do not really know what I am talking about here, but the bottom line is that it is a technical difficulty and they would like to implement it only if we think that it is important (I do, but what is your opinion?).
     
   • Francois Viallefond -- Archive
     
   • Chris Wilson -- Pipeline
     Since our July telecon I have
     1. drafted up expanded requirements for the Quick Look Pipeline; I haven't had time to look over Robert's comments on my new draft because of teaching an dother pressures in September
     2. created Use Cases for the Science Pipeline and Quick Look Pipeline (single field and mosaic, no single dish data included only)
     3. draft a test plan for the Pipeline subsystem and iterated on it once with the subsystem lead Lindsey Davis
     
     
   • Mel Wright -- Scheduling
     I reviewed the scheduling use cases written by Debra.
     We published Miriad and Gildas evaluations as BIMA and IRAM memos and the comparison and interoperability memos as ALMA memos.
     I made Miriad images from the Gildas data provided by Jerome.
   
   The need to keep enable the user to give phase II input before proposal evaluation (with phase I input) was briefly rediscussed. The SSR maintains its position that Phase II input should be possible at this stage if wanted. It is clear that this input should be validated but not necessarily by doing full phase II processing (i.e. building the whole program with its scheduling blocks).
   
   
2. News from ASAC Meeting (John Richer, Leonardo Testi)
   John Richer gave the SSR a brief summary of the very recent ASAC Meeting in Hamilton.
   
   
3. AIPS++ Test Phase III report updated (Robert Lucas)
   Final report is temporarily at http://www.iram.fr/%7Elucas/almassr/PhaseIIIfinal.pdf and will be submitted as an ALMA memo (together with Phase I and II reports). Thank to the constant efforts of the developpers, the current performance of AIPS++ on this test has greatly improved and is now well within a factor of 2 of Gildas performance (on machines with large memory).
   
   
4. U Off-Line Requirements (Debra Shepherd)
   Debra made a list of the these requirements (see her message of Tue, 12 Aug).
   • Some of the issues could be resolved based on work being done now (e.g. export data format)
   • On others we can't progress much at this time (e.g. operational issues still pending), list of supported platforms ...
   • on e.g. atmospheric/environmental parameters we can use, as a start, what is available at the ATF, and what is mentioned in ALMATI-FITS
   • The list or observing modes is not yet finalized
   • Debra will draft a list of possible DRP defects and the turn-around time that is expected for removing them.
   • We will go through this list every six months and see each time what points can be resolved.
   
   
   
5. Test Plans: status (Debra Shepherd)
   See Debra's report above.
   • Most subsystems have now provided Use Cases
   • Three subsystems have provided Test Plans; Gantt charts are available for OffLine and Pipeline. Debra offers to prepare Gantt charts for other subsystems, if the Subsystem Scientists provide the list of dates. Test plans should be made available by all subsystems in the course of the next month.
   • F. Viallefond will provide Archive Use Case(s) in the next few days.
   
   
   
6. Data Rates (Steve Scott, John Richer)
   Should a request for an increase be issued?
   We discussed the status of the data rates specification. John received comments from Larry D'Addario on his Science IPT note. He will forward them to the SSR list.
   Steve points out that our data rate will appear very conservative in 2012. He sent a message (after our July meeting) that did not raise any comment - please read it now and comment.
   We will carry this item on next month's agenda -- and should decide on action at that time.
   
   
7. Date and location of the face-to-face meeting
   It will be in Socorro on Nov. 14/15. The proposition is now to have the second meeting day at the ATF (Sat 15) and visit the prototypes at this occasion.
   
   
8. Can we have several correlator setups in one SB? (Leonardo Testi, Peter Schilke)
   (due to lack of time, this item is carried to the next meeting).
   Discussing with Peter Schilke, it came out that with the baseline correlator, a line survey project is highly inefficient, unless one manages to use several correlator setting (for a single tuning) in the same scheduling block. This, I believe, is not currently possible. Maybe we should discuss this point.
   
   
9. Date of next phone meeting The next one is on: 2003 October 8th 15:00 UT (still summer time in EU and US).