1. From the WEB to the GRID Industrial potential of the technology Fabrizio GAGLIARDI CERN Geneva-Switzerland EU-DataGrid Project Leader October 2001

2. Talk summary Introduction From the WEB to the Grid
EU DataGrid background
Future Plans
Potential for industry and commerce
Conclusions

3. From the WEB to the GRID
The history of computing is solutions in search of problems to solve
In the mid 80’s the problem of physicists at CERN was the exchange of multimedia information within international world-wide scientific collaborations

4. The WEB example All the elements of the solution were there:
Internet
Reasonably powerful PCs
Friendly user interfaces
Hypertext invented long before
Tim Berners-Lee in 1989 had the vision
A good invention which required to migrate to US to become a phenomenal success

5. Technological evolution
Networks: Qos, availability, cost
The Metcalf’s law: usefulness of networks grow with the cube of the number of their nodes
Internet exponential grow (traffic doubles every 12 months) PC
The Moore’s law: CPU power double every 18 months
User interfaces: Mosaic, Netscape, Portals

6. NSF PACI Network Connections

7. DataTAG project Abilene UK IT ESNET CERN GEANT MREN NL NewYork
SuperJANET4 IT
GARR-B
STAR-LIGHT
ESNET
GEANT
CERN
MREN NL
SURFnet
STAR-TAP

8. Asian Pacific Grid Common Framework for Asia-Pacific Grid researchers
Represent AP interests to GGF
Collaborate with APAN/TransPAC
Voluntary framework: Not a project funded from single source
North America
(STARTAP)

9. New step in technology
Wide area networking becoming as powerful, as reliable and affordable as local area networks
A PC today has the power of a computer center of “only” 10 years ago
Powerful graphics and friendly interfaces make access to computer resources very easy
In short: time ripe for a new vision

10. The CERN problem

11. The European Organisation for Nuclear Research
20 European countries
2,500 staff
6,000 users
CERN

12. 27km of tunnel stuffed with magnets and klystrons

13. One of the four LHC detectors
40 MHz (40 TB/sec)
online system
multi-level trigger
filter out background
reduce data volume
level 1 - special hardware
75 KHz (75 GB/sec)
level 2 - embedded processors
5 KHz (5 GB/sec)
level 3 - PCs
100 Hz
(100 MB/sec)
data recording &
offline analysis

14. The LHC Detectors
CMS
ATLAS
~6-8 PetaBytes / year
~108 events/year
LHCb

15. Funding Requirements growing faster than Moore’s law
CERN’s overall budget is fixed
Estimated cost of facility at CERN
~ 30% of offline requirements*
Budget level in 2000 for all physics data handling
R&D testbed
Physics WAN
Systems administration
Mass Storage
disks
processors
*assumes physics in July 2005, rapid ramp-up of luminosity

16. World Wide Collaboration  distributed computing & storage capacity
CMS: 1800 physicists 150 institutes 32 countries

17. LHC Computing Model Lab m Uni x Uni a Lab a Tier 1 Uni n Tier2 Lab b
Physics
Department   
Desktop
Tier2
Lab a
Uni a
Lab c
Uni n
Lab m
Lab b
Uni b
Uni y
Uni x
USA
Brookhaven
Tier 1
USA
FermiLab UK
France
Italy NL
CERN
……….
Germany

18. The solution: the GRID

19. The GRID metaphor Analogous with the electrical power grid
Unlimited ubiquitous distributed computing
Transparent access to multi peta byte distributed data bases
Easy to plug in
Hidden complexity of the infrastructure
The GRID metaphor has come into popular use recently. It is analogous with the electrical power grid, which provides transparent access to electrical power on demand. The computing grid provides transparent access to computing power on demand.
The GRID will hide the complexity of the infrastructure and how the resources are provided. Users will define their requirements in terms of databases, CPU, the GRID will look after the rest.
HEP requires high throughput rather that high performance. High throughput can be provided by many distributed systems.The GRID, which will provide access to remote computing resources is ideally suited to LHC computing. LHC computing is an ideal testbed for GRID technology
So, in some way Grid technology makes it easy to use diverse, geographically distributed, locally managed and controlled computing facilities – as if they formed a coherent local cluster
The LHC RC model is a typical GRID application combining computational and data grids.
Ian Foster and Carl Kesselman, editors, “The Grid: Blueprint for a New Computing Infrastructure,” Morgan Kaufmann, 1999,

20. EU DataGrid background
Motivated by the challenge of the LHC computing
Large amount of data (~10 Pbytes/year starting in 2006)
Distributed computing resources and skills
Geographical worldwide distributed community (VO)
Excellent Grid computing model match to HEP requirements (Foster’s quote: HEP is Grid computing “par excellence” )
Transition from supercomputers to commodity computing done
Distributed job level parallelism (no strong need for MPI)
High throughput computing rather than supercomputing
VO tradition already long established
Prototype Grid activity in some CERN member states

21. Main project goals and characteristics
To build a significant prototype of the LHC computing model
To collaborate with and complement other European and US projects
To develop a sustainable computing model applicable to other sciences and industry: biology, earth observation etc.
Specific project objectives:
Middleware for fabric & Grid management (mostly funded by the EU): evaluation, test, and integration of existing M/W S/W and research and development of new S/W as appropriate
Large scale testbed (mostly funded by the partners)
Production quality demonstrations (partially funded by the EU)
Open source and communication:
Global GRID Forum
Industry and Research Forum

22. Main Partners CERN – International (Switzerland/France) CNRS - France
ESA/ESRIN – International (Italy)
INFN - Italy
NIKHEF – The Netherlands
PPARC - UK

23. Research and Academic Institutes
Associated Partners
Research and Academic Institutes
CESNET (Czech Republic)
Commissariat à l'énergie atomique (CEA) – France
Computer and Automation Research Institute,  Hungarian Academy of Sciences (MTA SZTAKI)
Consiglio Nazionale delle Ricerche (Italy)
Helsinki Institute of Physics – Finland
Institut de Fisica d'Altes Energies (IFAE) - Spain
Istituto Trentino di Cultura (IRST) – Italy
Konrad-Zuse-Zentrum für Informationstechnik Berlin - Germany
Royal Netherlands Meteorological Institute (KNMI)
Ruprecht-Karls-Universität Heidelberg - Germany
Stichting Academisch Rekencentrum Amsterdam (SARA) – Netherlands
Swedish Natural Science Research Council (NFR) - Sweden
Industrial Partners
Datamat (Italy)
IBM (UK)
CS-SI (France)

24. More info on www.eu-datagrid.org
Project scope
9.8 M Euros EU funding over 3 years
90% for middleware and applications (HEP, EO and biology)
Three year phased developments & demos ( )
Possible extensions (time and funds) on the basis of first successful results:
DataTAG ( )
CrossGrid ( )
GridStart ( ) …
More info on

25. Potential for industry and commerce
New business model (open source + added value services)
Endorsed by three DataGrid partners
IBM recent announcements and plans
Integration and service providers
Opportunity for ASPs
Electronic commerce enabler

26. Few industrial examples
NASA: for on-line diagnostic
Boeing: HPC simulation for engineering design
ESA: several EO compute and data intensive applications
VC exploring other business opportunities (see Index Venture presentations at GGF3 in Frascati)

27. Few scientific examples

28. What we will be able to do If Grids and Networks continue to grow
ONLINE ANALYSIS OF
INSTRUMENT DATA
TELE-IMMERSION/DISTANCE
COLLABORATION
TRANS-ATLANTIC REMOTE VISUALIZATION AND STEERING
COLLABORATIVE DATA MINING
RECORD-SETTING DISTRIBUTED
SUPERCOMPUTING

29. Example Application: Online Instrumentation
real-time
collection
wide-area
dissemination
desktop & VR clients with shared controls
Advanced Photon Source
archival
storage
tomographic reconstruction
DOE X-ray source grand challenge: ANL, USC/ISI, NIST, U.Chicago

32. Improving Severe Storm Forecasting: Using the Grid to Gather the Initial Data
Upper-Air
Balloons
Satellites
NEXRAD
Doppler
Radar
Commercial Aircraft
Automated
Surface
Networks

33. Conclusions
EU DataGrid is well on its way to demonstrate that Grid is the right solutions for CERN and LHC computing
The intense flourishing of Grid projects in other disciplines demonstrates that Grid is good for science
I believe that industry and commerce will be next, provided we manage to build secure Grids with internationally accepted standards
The Global Grid Forum recently launched should contribute to this process (