Meeting Goals

1
Meeting Goals

• review achievements
• review architecture
• identify targets for AG2 and VOTECH
• gaps
• technology changes
• components for re-engineering

2
Tuesday

• AM overviews / guest talks
• LUNCH 1230-1330
• PM Demos / Tech detail / planning
• DINNER Beluga, Chambers St 1930

3
Dinner
NeSC
4
Wednesday

• 0930 start
• Architecture review and analysis
• 1530 finish

5
(No Transcript)
6
State of the Union

• AstroGrid scorecard
• The global VO
• AstroGrid today

7
AstroGrid scorecard
8
achievements since april meeting

• Integration
• Deployment
• Architecture
• VOTECH funded
• kick-off meeting November

9
AstroGrid timeline

• Apr 01 Proposal to PPARC

10
words from april 2001 proposal
(3.1) LONG TERM PERSPECTIVE The vision of a
"Virtual Observatory" only makes eventual sense
as a global entity (GVO ?)..... It is our
impression that such aspirations will only become
reality on a timescale of perhaps 6 years or so,
but that the components and toolkits will start
to develop much sooner. Likewise, the underlying
computer science will on the one hand evolve as
we try to apply it, and on the other hand will be
looking for challenging applications to influence
its evolution. In this perspective we believe
that a major but short-term (three year) project
is the correct first step, as after this we must
take stock and re-group. The UK cannot build the
GVO by itself, but we should not wait either. Our
overall aims are therefore (i) to provide a
working system for UK astronomers to exploit key
databases, (ii) to develop tools and methods
that the international community will see as key
contributions to an eventual GVO, and (iii) to
learn technical lessons as fast as possible that
will influence the design of an eventual GVO.
11
AstroGrid timeline

• Apr 01 Proposal to PPARC

12
AstroGrid timeline

• Apr 01 Proposal to PPARC
• Oct 01 - Dec 02 Phase A study
• Jan 03 - Dec 04 Phase B construction-------------
  ------------------------------------
• Jan 05 - Dec 07 AstroGrid2

13
AstroGrid timeline

• Oct 01 - Dec 02 Phase A study
• Technology assessment
• Jan 03 - Dec 04 Phase B construction
• Infrastructure components s/w beta release
• First deployment beta tester release
• Jan 05 - Dec 06 AstroGrid2
• re-engineering
• more compts
• multiple deployments
• next wave RD (VOTECH)

14
Promises to PPARC Phase B goals

• the eightfold path ...

15
goal check (1-4)

• develop standards for data, metadata, data
  exchange and provenancedone with IVOA
• develop a software infrastructure for data
  servicesdone but still growing
• establish a physical grid of resources done with
  DC help modest start
• construct and maintain a Service and Resource
  Registrydone

16
goal check (5-8)

• implement a working VO system of real scientific
  use to astronomershalf done beta test stage
• provide a user interface to that VO
  systemtechnical components doneuser friendly
  skin half done
• provide or adapt a set of science user tools to
  work with the VOprovide officially abandoned
  due to lack of resourceadapt done
• establish a leading position for the UK in VO
  workdone

17
the global VO
18
the global VO
key AstroGrid contributions
19
publishing metaphor

• experiments/facilities are authors
• data centres are publishers
• VO portals are shops
• end-users are readers
• VO infrastructure is distribution system.

20
work to do

• agree standards
• build VO-aware data services
• build VO-aware tools
• build software glueconstruct deployments, get
  users

21
progress

• agree standards
• excellent progress through IVOA
• build VO-aware data services
• build VO-aware tools
• build software glue

22
International VO alliance (IVOA)
http//www.ivoa.net
23
progress

• agree standards
• excellent progress through IVOA
• build VO-aware data services
• data centres upgrading worldwide
• build VO-aware tools
• a few examples CEA should allow to take off
• build software glue
• Registry, Workflow, VOspace...
• sign-on to come

24
deployments and users

• several hand crafted-demos
• some real science done
• US-NVO has some simple production websites
• OpenSkyQuery / DataScope / YourSky
• AstroGrid portal will allow more complex jobs
• first users starting to play
• AVO SWG AstroGrid AGSAG
• NVO summer school

25
does it look like a VO yet ?

• data interoperability
• resource interoperability
• data integration
• single sign on
• complex jobs

26
does it look like a VO yet ?

• data interoperability
• good progress
• data access VOtable, SIAP, SSAP
• semantic dictionary UCDs
• will need refreshing

27
does it look like a VO yet ?

• resource interoperability
• good progress
• Resource and Service Metadata
• harvestable registries
• standard query language ADQL
• progress on WS interfaces
• storage sharing (MySpace)

28
does it look like a VO yet ?

• data integration
• not really there only plain cone-searches
• really science / algorithm problem not IT
• need real semantics - not just types
• UCNs units grammar
• concentrate on operations not philosophy

29
catalogue xmatch is the easy bit ...
30
federating these is a bit more subtle
Shocks seen in the X-ray
Dust seen in the IR
Heavy elements seen in the optical
Relativistic electrons seen in the radio
31
as is these...
Effect detected hours later by satellites and
ground radar
Coronal mass ejection imaged by space-based solar
observatory
32
does it look like a VO yet ?

• data integration
• not really there only plain cone-searches
• really science / algorithm problem not IT
• need real semantics - not just types
• UCNs units grammar
• concentrate on operations not philosophy

33
does it look like a VO yet ?

• single sign on
• limited IVOA attention (GWS-WG)
• AstroGrid shell in place
• need to cohere with rest of academia
• Shibboleth looks like the winner ?

34
does it look like a VO yet ?

• complex jobs workflow
• negligible IVOA attention
• AstroGrid workflow engine and UI
• AstroGrid application integration

35
AstroGrid today
36
AstroGrid strands

• Standards development
• RD
• Infrastructure software product
• Deployment

37
Customers

• Standards other VO projects
• RD ourselves, other VO projects
• Software DCs, VO projects, tools writers
• Deployment astronomical community

38
Software product aims

• product quality
• mix-and-match
• plug-and-play
• publisher help (PAL and CEA)

39
Secret Weapons

• MySpace
• Workflow
• Application integration CEA

40
Deployment today

• Registry, Workflow, MySpace, Portal
• Distributed data, storage, and services
• Leicester, Geneva, MSSL, Edinburgh, Cambridge,
  ESO
• Querying, xmatching, cpu-intensive analysis
• VO-aware applications
• e.g. Sextractor, HyperZ, moviemaker

41
Issues

• Stability
• Useability
• Missing components
• IAA, scripting, workbench, grid services
• Improved functionality
• intelligent resource discovery
• ambitious data mining services
• using GridPower

42
FIN