Tags:
create new tag
view all tags

Consortium Meeting Minutes: 06 October 2005

Location

  • Date and time: Thursday, 6 October 2005, 10:30 - 16:00
  • Location: Hicks Building, Room G12, University of Sheffield
  • Parking: Email Lynda Harrison by Sept. 29 to reserve a parking space on campus
  • Map: campus map and map of Sheffield city centre. To find the Hicks Building, look at the campus map. Starting from the Crookes Valley Park boating lake, look to the right down Winter Street and Bolsover Street to the big roundabout. Look down Upper Hanover Street past Leavygreave Road, and the Hicks Building is on Hounsfield Road to the left. The Hicks Building is also marked on the city centre map.

Agenda

<10:30 Welcome, Coffee EA
10:30 Phase A Review - deliverables, PPARC review panel, UK solar community EA
11:00 Algorithms and discussion TT, SZ, VG, MS, EA
12:45 Lunch all
13:30 Data Centres - US and UK EA
14:15 Quicklook and Visualization EA
14:30 HELAS MT
14:45 Phase B: workpackages EA, discussion
15:15 Phase B: technical information EA
15:30 AOB all
16:00 finish  

Phase A Review

The PPARC review of eSDO Phase A will take place on 28 November 2005 at the MRC. Panel: Declan Diver (Glasgow), Stephen Tobias (Leeds), Gerry Doyle (Armagh). eSDO attendees: Elizabeth, Len, Mike, Yvonne(?), Andrzej(?). There will be 4 presentations: 1) Intro to eSDO, AIA and HMI (Len and Mike), 2) Phase A (Elizabeth), 3) Phase B work plans (Elizabeth), and 4) Phase B budget and technical detail (if required, Elizabeth).

Algorithms

See developer presentations attached below. Most algorithms have been identified as computationally intense, and this should be reflected in the post launch support proposal for algorithms that will be hosted as AstroGrid services within the UK. Many of the 11 original algorithms will be more sensibly packaged as single algorithms, leaving 8 algorithms:

  1. Mode parameters and assymetry
  2. Magnetic field extrapolation and helicity computation
  3. Perturbation map generation and subsurface flow analysis
  4. Local helioseismology inversion
  5. Coronal loop recognition
  6. Small event detection
  7. CME dimming region recognition
  8. DEM computation

Data Centres

The 30 TB "light footprint" model for a UK data centre was agreed as the first choice of architecture. See Elizabeth's consortium meeting presentation below for more details.

Quicklook and Visualization

Thumbnail extracation, image gallery generation and movie generation techniques were agreed. There is support for the SDO streaming tool, although research into open source licenses versus commercial possibilities will need to be investigated carefully. Workpackages for rapid prototyping should be agreed with US colleagues before Phase A review. Quicklook catalogues vs dataset catalogues should be revisited with Mike and Yvonne. See Elizabeth's presentation for more details.

HELAS

Mike Thompson discussed developments with HELAS, a European consortium that will work together on helioseismology analysis. The consortium's interests will overlap with eSDO algorithm development at Sheffield and Birmingham; both groups should find synchronicities between eSDO algorithm outout and HELAS algorithm inputs. There will be opportunities for eSDO project members to attend HELAS consortium meetings. Plans for a HELAS SDO data centre are unknown to eSDO at this time.

Phase B

Developers will have fortnightly telecons to keep up with progress. Developers will begin algorithm coding throughout October; by the end of the month, procedures for CVS, CUtest unit testing and backups will be in place. All developers will plan to attend the AIA / HMI workshop in February 2006.

Actions

  1. ALL - Examine intensive computing needs for eSDO algorithms hosted in the UK that will need to be scoped for the post launch support proposal.
  2. ALL - consider potential questions from PPARC Phase A review panel regarding the ability to write intensive computing algorithms plus visualization and data centre applications within time and budget.
  3. ALL - start coding algorithms.
  4. ELIZABETH - liase with UCL intellectual property advisors regarding open sources licenses and commercial opportunities with the SDO streaming tool.
  5. ELIZABETH - amend QuicklookVisualization1211 to form two seperate catalogues for small event detection and CME dimming region recognition catalogues.
  6. ELIZABETH - distribute the Phase A report URLs to further identified members of the solar and grid communities.
  7. ELIZABETH - add HMI LoS magnetograms and continuum maps to the list of systematically cached data products to be held in UK data centre for 60 days.
  8. ELIZABETH - investigate open source licese for eSDO code, shared code with NASA, and SDO streaming tool.
  9. ELIZABETH - liase with Mike T. and Sergei about local helioseismology catalogue (quicklook vs dataset)
  10. ELIZABETH - discuss ad hoc catalogues with Mike T. and Yvonne.
  11. ELIZABETH - liase with Yvonne to find suitable date for visit to Birmingham
  12. LEN - Talk to Deborah Miller about 1) whether members of all eSDO institutions should be present for Phase A review, and 2) if summary report or full Phase A report is more appropriate for panel members.
  13. LEN / MIKE T. - speak to Sami and Laurent about further eSDO interaction with HELAS.
  14. MIKE S. - investigate SQL syntax required to return MySQL fields in specific formats (especially dates, doubles). Also look into ADQL syntax or CEA application that will convert between the two.
  15. MIKE T. - liase with HELAS co-investigators regarding developmental overlap between eSDO and HELAS algorithms. Also investigate ensuring eSDO outputs and HELAS inputs are of compatible formats.

-- ElizabethAuden - 10 Oct 2005

Topic attachments
I Attachment History Action Size Date Who Comment
PowerPointppt ConsortiumMeeting061005.ppt r1 manage 419.0 K 2005-10-10 - 17:28 ElizabethAuden Elizabeth's consortium meeting presentation
PowerPointppt Frequencyandasymmetry.ppt r1 manage 825.0 K 2005-10-10 - 17:29 ElizabethAuden Thierry's algorithm presentation
PowerPointppt RAL-Algorithms.ppt r1 manage 1932.0 K 2005-10-10 - 17:30 ElizabethAuden Vito's algorithm presentation
PowerPointppt Sheffield06102005_V1.ppt r1 manage 441.0 K 2005-10-10 - 17:31 ElizabethAuden Sergei's algorithm presentation
Quicktime moviempeg artificial_dopplergrams.mpeg r1 manage 468.6 K 2005-10-10 - 17:32 ElizabethAuden Thierry's movie of artificial dopplergrams
Microsoft Word filedoc mssl_algorithms.doc r1 manage 43.5 K 2005-10-10 - 17:31 ElizabethAuden Mike S.'s algorithm presentation
Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | More topic actions
Topic revision: r2 - 2005-10-10 - ElizabethAuden
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback