1 / 82

The Future of GRIDs: A European Perspective

The Future of GRIDs: A European Perspective. Keith G Jeffery Science and Technology Facilities Council Rutherford Appleton Laboratory, OX11 0QX UK e-mail: keith.g.jeffery@rl.ac.uk. Old job – running the major IT department Computer operations 360,000 users 1100 servers

andres
Télécharger la présentation

The Future of GRIDs: A European Perspective

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. The Future of GRIDs: A European Perspective Keith G Jeffery Science and Technology Facilities Council Rutherford Appleton Laboratory, OX11 0QX UK e-mail: keith.g.jeffery@rl.ac.uk

  2. Old job – running the major IT department Computer operations 360,000 users 1100 servers Systems development Corporate Departments External R&D Library and Information Services Photoreprographics Director, IT & International Strategy Who ?

  3. And… • President ERCIM • President euroCRIS • Honorary Visiting Professor • University of Cardiff • Heriot Watt University Edinburgh • Masaryk University Czech Republic • Fellow BCS and GS, Honorary Fellow ICS, CEng, CITP

  4. CCLRC-RAL Site

  5. PPD: CERN: LHC

  6. PPD : CMS

  7. ISIS: Neutrons

  8. SSTD: Earth Gulf Stream Etna

  9. SSTD: Mars • 23 January 2004This picture was taken by the High Resolution Stereo Camera (HRSC) onboard ESA's Mars Express orbiter, in colour and 3D, in orbit 18 on 15 January 2004 from a height of 273 km. The location is east of the Hellas basin at 41° South and 101° East. The area is 100 km across, with a resolution of 12 m per pixel, and shows a channel (Reull Vallis) once formed by flowing water. The landscape is seen in a vertical view, North is at the top.

  10. SSTD: Earth Dartford, UK (with inset of the Queen Elisabeth Bridge) from 680km up First image from RAL camera on TOP-SAT

  11. Lasers: Vulcan

  12. Diamond: Synchrotron Radiation

  13. Computing • STFC runs HPC(X) • 5th fastest computer when purchased • IBM Power series • Used by UK R&D Community

  14. VR: EISCAT Control Problem: cost and data loss when training scientists to use EISCAT Answer: VR system at RAL to train before going to Norway

  15. ERCIMEuropean Research Consortiumfor Informatics and Mathematics • 18 European countries - major labs or consortia of universities • 12000 ICT researchers • Working groups • Fellows programme • Cor Baayen Award • Strategy documents for EC and national governments • R&D projects, networks of excellence etc • > 100 spin-out companies • Host of W3C Europe • European Office(s) of W3C • ERCIM News • www.ercim.org

  16. Linking together systems in each country managing research information • Funders of research • Organisations performing research • For • Strategic decision-making about ewhat research to fund /do • Finding research partners and competitors • Finding innovative ideas for technology transfer / exploitation • Informing the media / public • www.eurocris.org CERIF: an EU Recommendation to member states

  17. So? • This background gives you some idea of ‘where I’m coming from’ • Advanced research problems requiring ICT solutions • Research not ‘blue sky’ but practical • Management and administrative systems for Research Support • International working • Strategic thinking for / using blue sky research to plan roadmaps for ICT R&D

  18. So? • This background gives you some idea of ‘where I’m coming from’ • Advanced research problems requiring ICT solutions • Research not ‘blue sky’ but practical • International working • Strategic thinking for / using blue sky research to plan roadmaps for ICT R&D • And what I am going to talk about is the ICT of the future that we shall all be using and/or developing • And the research challenges we have to overcome to make it happen

  19. STRUCTURE • The Original UK Idea • Where We Are Now • The R&D Required to Achieve GRIDs • NGG: Next Generation GRIDs • CoreGRID • Challengers • Conclusion

  20. In the beginning….. • In 1999 the UK Research Councils (which fund university R&D) were undergoing their Strategic Review Exercise for funding beyond 2000 • Grand challenge science projects • The DGRC (John Taylor) unhappy that plans • had too little IT • the IT proposed was incoherent • So he asked CCLRC CEO (Bert Westwood) to have someone generate an IT plan • And Bert asked me

  21. The GRIDs Vision • The end-user interacts with the GRIDs environment to clarify the request • using a ‘device’ or ‘appliance’ • The GRIDs environment proposes a ‘deal’ to satisfy the request • which may or may not involve money • The user accepts or rejects the ‘deal’

  22. The GRIDs Vision • The GRIDs environment is such that • A user can interact with it intelligently • It provides transparent access to • data, information, knowledge • computation • instrumentation / detectors

  23. The GRID Bible

  24. Knowledge Layer Information Layer Data toKnowledge Control Computation / Data Layer The GRIDs Architecture

  25. Data toKnowledge Control Particle Physics Application Genomics Application Environmental Application E-Business Application The GRIDs Architecture

  26. The GRIDs Environment User Appliance Plug-in Personal Communication Personal Shopping Hobbies, family activities Business Communication Business Dealing Business Information PC Palmtop Mobile.. The Big Idea:What it Provides ‘The Wall’

  27. Ambient, pervasive, mobile • The user appliance may well be mobile and requires pervasive connectivity • It may have interesting capabilities such as attachment of detectors / instruments • Scientific research • paramedics, firefighters • Even ‘road warriors’

  28. U:USER R:RESOURCE S:SOURCE A POSSIBLE ARCHITECTURE The GRIDs Environment Um:User Metadata Ua:User Agent Sm:Source Metadata Sa:Source Agent Ra:Resource Agent Rm:Resource Metadata brokers

  29. Classification of Metadata view to users SCHEMA NAVIGATIONAL ASSOCIATIVE constrain it how to get it data (document)

  30. Representative Agents • Represent the entities {U, S, R} continuously and actively within the GRIDs environment • With metadata represent the entity to others represented by their agents • Act on behalf of the entity

  31. Brokers • (a) authentication, • (b) clarification / precision of request, • (c) resource discovery (information and if necessary compute power, visualisation facilities etc) • (d) authorisation (rights), • (e) offer and pricing, • (f) closure of deal (U accepts (e))

  32. Brokers (continued) • (g) fusion of responses, • (h) application of any transformation / analysis / simulation / visualisation processes, • (i) presentation formatting (for variously abled devices and people using various resources), • (j) network routing, and (k) scheduling of physical resource access / usage

  33. Monitoring Brokers • and others will monitor • quality of service, • utilisation of resource collections • specialist physical resources • etc etc.

  34. STRUCTURE • The Original UK Idea • Where We Are Now • The R&D Required to Achieve GRIDs • NGG: Next Generation GRIDs • CoreGRID • Challengers • Conclusions

  35. A Brief History of GRIDs • 1G: custom-made architecture machines to user • Pioneering metacomputing • 2G: proprietary standards and interfaces • I-WAY GLOBUS, UNICORE, CONDOR, LEGION AVAKI • 2.5G: added in FTP, SRB, LDAP, AccessGRID • 3G: adopted W3C concepts for open interfaces – OGSA / OGSI: note especially OGSA/DAI • But built on 2.G foundations

  36. A Brief History of GRIDs • 1G: custom-made architecture machines to user • Pioneering metacomputing • 2G: proprietary standards and interfaces • I-WAY GLOBUS, UNICORE, CONDOR, LEGION AVAKI • 2.5G: added in FTP, SRB, LDAP, AccessGRID • 3G: adopted W3C concepts for open interfaces – OGSA / OGSI: note especially OGSA/DAI • But built on 2.G foundations e-Science Apps

  37. A Brief History of GRIDs • 1G: custom-made architecture machines to user • Pioneering metacomputing • 2G: proprietary standards and interfaces • I-WAY GLOBUS, UNICORE, CONDOR, LEGION AVAKI • 2.5G: added in FTP, SRB, LDAP, AccessGRID • 3G: adopted W3C concepts for open interfaces – OGSA / OGSI: note especially OGSA/DAI • But built on 2.G foundations e-Science Apps e-Science R&D

  38. But….. • This comes nowhere near the requirements as originally defined for GRIDs • Too low-level (programmer not end-user level) • Insufficient representativity • Insufficient expressivity • Insufficient resilience • Insufficient dynamic flexibility

  39. STRUCTURE • The Original UK Idea • Where We Are Now • The R&D Required to Achieve GRIDs • NGG: Next Generation GRIDs • CoreGRID • Challengers • Conclusions

  40. 1999-2000 • The R&D issues were proposed in late 1999 • Discussed and approved at a meeting in UK of representative ‘gurus’ from academia and industry in early 2000 • Essential Technologies needing R&D • Ease of use • Trust • Performance

  41. Facilitate Ease of Use • Metadata • Representation language expressivity: user, source, resource • within / across domains • Agents • Specialised or Generalised and configured by metadata • Dynamically reconfigured by events / messages • Brokers • Functional • Knowledge-based with some autonomy • Strategic knowledge

  42. Facilitate Trust • Security • Wireless communications • Availability of service • Privacy • Tradeoff personal information for intelligent system reaction • Trust • Of services / servers • Of payment for services

  43. Facilitate Performance • Mobile code • Be able to move the code to the data rather than data to the code • Share code among nodes active in one request • Performance and optimisation • Synchronisation, consistency, reliability • Ease of management

  44. STRUCTURE • The Original UK Idea • Where We Are Now • The R&D Required to Achieve GRIDs • NGG: Next Generation GRIDs • CoreGRID • Challengers • Conclusions

  45. So….. • The US GRID is metacomputing plus extensions • In 2002 improved with OGSA using W3C Web Services ideas • European position is that GRID architecture (GLOBUS or even UNICORE) is the wrong starting point for the European vision

  46. And….. • EC persuaded of importance of GRIDs • Started in IST/Environment (early 2000) with IT architectural framework for FP6 projects • Set up GRID Unit under Wolfgang Boch (late 2002) • January 2003: large workshop (GRID Unit) • (~ 240 participants) • Keynotes: • Thierry Priol (INRIA, FR) • Domenico Laforenza (CNR, IT) • Keith Jeffery (CCLRC, UK)

  47. NGG Requirements • Transparent and reliable • Open to wide user and provider communities • Pervasive and ubiquitous • Secure and provide trust across multiple administrative domains • Easy to use and to program • Persistent • Based on standards for software and protocols • Person-centric • Scalable • Easy to configure and manage

  48. Call2 (NGG1) Projects Funded

  49. NGG2 • NGG1 left some undefined research areas • Call2 projects did not address all areas of research opportunity • NGG2 convened to update the vision: • Particularly security / trust • Particularly self-* properties • Particularly semantic description of components • Report September 2004

  50. Application A Application B Application C Grids Middleware Services Needed for A Grids Middleware Services Needed for B Grids Middleware Services Needed for C Grids Foundations for Operating System X Grids Foundations For Operating System Y Grids Operating System (including Foundations) Modular and dynamically loadable Operating System X Operating System Y NGG2 Architecture

More Related