1. A sustainable e-Infrastructure for Europe Athens, Greece 19 April 2006 Bob Jones

2. Athens, 19 April 2006 2 Contents The vision of an e-Infrastructure for Europe The international perspective Proposing a possible model –Mission –Structure –Key services –Funding & Governance Discussion Additional material – details of key services

3. Athens, 19 April 2006 3 e-Infrastructure for Europe The Vision (1) –“An environment where research resources (H/W, S/W & content) can be readily shared and accessed wherever this is necessary to promote better and more effective research” (1) “A European vision for a Universal e-Infrastructure for Research” by Malcolm Read http://www.e- irg.org/meetings/2005-UK/A_European_vision_for_a_Universal_e-Infrastructure_for_Research.pdfhttp://www.e- irg.org/meetings/2005-UK/A_European_vision_for_a_Universal_e-Infrastructure_for_Research.pdf

4. Athens, 19 April 2006 4 International Perspective Recommendations (1) of the Global Science Forum (2) of the Organisation for Economic Co-operation and Development 1: “Grids deserve to be treated as research infrastructures in their own right….. Governments should also consider taking steps to strengthen the international mechanisms for co-operation and co-ordination at the scientific, commercial and intergovernmental levels” 2: “A number of important Grid-related issues need to be addressed in a strategic, systematic, organised, international manner, with the participation of all the stakeholders: funding agency officials, Grid architects, implementers, and users from the scientific, academic and industrial communities” 3: “Consideration should be given to the creation of new mechanisms (or the strengthening of existing ones) to facilitate access to Grids for researchers and research organisations in developing countries, plus other appropriate measures to broaden international participation in grid projects” (1) Report on Grids and Basic research Programmes, Sydney September 25-27 2005, http://www.oecd.org/dataoecd/30/36/36213997.pdfhttp://www.oecd.org/dataoecd/30/36/36213997.pdf (2) Australia, Austria, Belgium, Denmark, France, Germany, Italy, Japan, Netherlands, New Zealand, Sweden, United Kingdom, United States

5. Athens, 19 April 2006 5 The view of the e-IRG e-IRG Recommendation: “The e-IRG recognises that the current project-based financing model of grids (e.g., EGEE, DEISA) presents continuity and interoperability problems, and that new financing and governance models need to be explored – taking into account the role of national grid initiatives as recommended in the Luxembourg e-IRG meeting.” White Paper: http://www.e-irg.org/publ/2005-Luxembourg-eIRG-whitepaper.pdfhttp://www.e-irg.org/publ/2005-Luxembourg-eIRG-whitepaper.pdf

6. Athens, 19 April 2006 6 Proposed model of a e-Infrastructure for Europe This model builds on the experience gained with the EGEE project, infrastructure and applications to define a European Grid Infrastructure (EGI) Takes into account the OECD & eIRG recommendations Input from EGO paper (1) and the workshops –Thoiry (France) 30-31 st January – EGEE federations –Kassel (Germany) 10 th March – Germany –Barcelona (Spain) 28 th March – France, Portugal, Spain Further planned events –Athens (Greece) 19 th April – Bulgaria, Cyprus, Greece, Israel, Romania, Serbia, Turkey –Vilnius (Lithuania) 26 th April – Estonia, Finland, Latvia, Lithuania, Poland, Sweden –Paris (France) 28 th April – Terena/NRENS grid workshop (1) Establishing an European Grid Organisation (EGO), http://www.e-irg.org/meetings/2005-UK/050617-EGO-position-paper.pdf http://www.e-irg.org/meetings/2005-UK/050617-EGO-position-paper.pdf

7. Athens, 19 April 2006 7 Mission Infrastructure –Manage and operate production Grid for European Research Area –Interoperate with e-Infrastructure projects around the globe –Contribute to Grid standardisation and policy efforts Support applications from diverse communities –Astrophysics –Computational Chemistry –Earth Sciences –Finance –Fusion –Geophysics –High Energy Physics –Life Sciences –Multimedia –… Business –Forge links with the full spectrum of interested business partners Disseminate knowledge about the Grid through training

8. Athens, 19 April 2006 8 Structure Federated model bringing together National Grid Initiatives (NGIs) to build a European organisation EGEE-II federations would evolve into NGIs Each NGI is a national body Recognised at the national level Mobilises national funding and resources Contributes and adheres to international standards and policies Operates the national e-Infrastructure Application independent, open to new user communities and resource providers

9. Athens, 19 April 2006 9 European National Grid Projects Austria – AustrianGrid Belgium – BEGrid Bulgaria – BgGrid Croatia – CRO-GRID Cyprus – CyGrid Czech Republic- METACentre Slovakia Denmark Finland France – planned (ICAR) Germany – D-GRID Greece - HellasGrid Hungary Ireland - Grid-Ireland Israel – Israel Academic Grid Italy - planned Netherlands – DutchGrid Norway – NorGird Poland Portugal - planned Romania – RoGrid Serbia – AEGIS Slovenia SiGNET Spain – planned (IBERgrid) Sweden – SweGrid Switzerland Turkey – TR-Grid United Kingdom - eScience

10. Athens, 19 April 2006 10 A global, federated e-Infrastructure EGEE infrastructure ~ 200 sites in 39 countries ~ 20 000 CPUs > 5 PB storage > 10 000 concurrent jobs per day > 60 Virtual Organisations EUIndiaGrid EUMedGrid SEE-GRID EELA BalticGrid EUChinaGrid OSG NAREGI

11. Athens, 19 April 2006 11 EGI Key Services Based on experience gathered during EGEE, the following key services have been found necessary for a central organisation in coordination with the NGIs –The information on the following slides are taken from existing EGEE structures and procedures as a means of explaining the concepts Operation of Infrastructure  Runs Operational Coordination Centre linking a Regional Operations Centre (Point of Presence) in each NGI  Coordinates the grid security and resource accounting  Negotiates resources for user communities  General management of the User Support process  Interaction point with GEANT for grid issues  Provides documentation for end users and resource providers

12. Athens, 19 April 2006 12 EGI Key Services (cont) Middleware testing and certification  Integrates middleware from other sources to produce distributions  Provides first-level middleware support team  Operates beta-test services for upcoming distributions Application support  No direct support but rather coordination of NGI support groups and management of overall application lifecycle (virtuous cycle) Dissemination and outreach  Branding, media relations, production of promotional material, websites etc  Event organisation  Public outreach & surveys - Ensuring higher level of media coverage as technology matures and becomes available to more end users  Representation of NGIs in international bodies and standards groups at a policy level Training  Aid in the formation of NGIs and their management/operations  Training material repository

13. Athens, 19 April 2006 13 Capability and Capacity Facilities The users want to see a unified e-Infrastructure providing access to both capability (super-computer centres) and capacity (PC cluster-based centres) –Several communities (Fusion, Life sciences etc.) have a need for both types of facilities and want to be able to move applications and data between them EGI will work with super-computer centres and these user communities to ensure they are accessible –EGI will not be responsible for the operations, funding or management of super-computer centres

14. Athens, 19 April 2006 14 Support for digital repositories GÉANT. INFRASTRUCTURE GRID. INFRASTRUCTURE KNOWLEDGE. INFRASTRUCTURE EGI will work with the digital library community to support access to heterogeneous information and connect resources through common shared services

15. Athens, 19 April 2006 15 EGI/NGIs and industry Business model for how industry can commercially exploit the research infrastructure managed by EGI/NGIs is unclear –Should not use tax payers money to compete with commercial service providers Likely to see transfer of technology from research to industry by adoption/internalisation of EGI/NGI backed products and services (e.g. middleware, operations procedures/techniques) –For multi-site corporate usage or to offer a service to a set of SMEs –Several examples already exist for EGEE EGI/NGIs could subcontract infrastructure support to industry and make use of commercial software as standards evolve

16. Athens, 19 April 2006 16 EGI Managed Resource Centres Given the existence of such an e-Infrastructure, EGI managed resources centres can be established: –Create shared pool of resources (CPU, disk and data curation) independent of funding for specific user communities –Joint capital funding from NGIs and EU as part of new resource infrastructure –Selection based on bids against a defined service level agreement –Business models with pay-per-usage to cover operational and depreciation costs –Would create/test example business models for potential future commercial supply and/or exploitation

17. Athens, 19 April 2006 17 EGI Governance and Funding Governance –Organisation with its own legal identity –NGIs are the stakeholders  NGIs would form the governing council  Annual reviews by independent experts nominated by the EU What to Fund –Basic infrastructure and its operation including national Points-of- Presence, regional resource centres and central organisation How to Fund –Basic funding by NGIs (50%) and EU (50%) ? –EU could fund preparatory project to set-up EGI through the New Research Infrastructures scheme (~12 months?) –Full EU co-funding (FP7) could start in 2008/9

18. Athens, 19 April 2006 18 The role of CERN in EGI CERN has a long-term interest in ensuring e- Infrastructures continue to develop to be able to support its LHC programme over the next 15 years CERN is heavily involved in the e-Infrastuctures development but hopes that it can become a client of such services in the future, as has been the case for networks The CERN Director General has offered to host the EGI organisation temporarily during its incubation

19. Athens, 19 April 2006 19 Links between infrastructure and R&D 1.Sustainable e-Infrastructure –Long-term cycle (up to 7 years) –Production/operations focus –Application independent –Open and inclusive 2.On-going developments –Further developments of middleware, applications, security etc. –Separate projects coordinated with the infrastructure –Limited duration (2 to 3 years) Complementary streams where successful developments (2) should become part of the sustainable infrastructure (1) Increase (2) →(1) by making the infrastructures stakeholders in R&D projects Possible combination of use of EU instruments for on-going and new infrastructures

20. Athens, 19 April 2006 20 Summary The need for a European e-Infrastructure has been identified The current structures are reaching their limits A model committing the National Grid Initiatives and building a central organisation is proposed – your input and feedback is actively sought Proposed key services –Operation of Infrastructure –Middleware testing and certification –Application support –Dissemination and outreach –Training Such a scheme will ensure a sustainable e-Infrastructure for research and help maintain Europe’s leading position

21. Athens, 19 April 2006 21 Details on EGI key services The following slides provide greater detail of the key services proposed for the central organisation in coordination with the NGIs These details are based on the experience gathered through EGEE and related projects

22. Athens, 19 April 2006 22 JSPG: Joint Security Policy Group OSCT: Operational Security Coordination Team Operations Coordination Regional Operations Centre … … Resource Centre Resource Centre … Resource Centre Resource Centre … Operations Coordination Centre OSCT Coordination, Middleware deployment Operational security coordination JSPG Coordination, Middleware deployment Coordination, Middleware deployment EGI NGI local

23. Athens, 19 April 2006 23 Regional Operations Centre (ROC) Distributed responsibility for operations: One ROC in each NGI –A grid “Point of presence” –Manage daily grid operations – oversight, troubleshooting  “Operator on Duty” with weekly rota between NGIs –Run infrastructure services (not applications themselves) –Support for user and operations issues –Provide regional knowledge and adaptations Operations coordination –Regular operations & managers meetings –Series of Operations Workshops Procedures described in Operations Manual –Introducing new sites –Site downtime scheduling –Suspending a site –Escalation procedures –Etc.

24. Athens, 19 April 2006 24 ROC and Site work to resolve the problem Regional Operations Centre …… Resource Centre Resource Centre … OSCT Grid Operator on-duty Monitoring shows a problem Operator submits a GGUS ticket against the ROC and cc’s the site. The ticket is followed until it is solved Operations support workflows Regional Operations Centre Resource Centre Resource Centre … EGI NGI Local support

25. Athens, 19 April 2006 25 Coordination of Grid Security Joint Security Policy Group –Across many Grid infrastructures –Policy Set: Policy Revisions Grid Acceptable Use Policy (AUP) –for all VO members using many Grid infrastructures  EGEE, OSG, SEE-GRID, DEISA, national Grids… VO Security –responsibilities for VO managers and members –VO AUP to tie members to Grid AUP accepted at registration Incident Handling and Response –defines basic communications paths –defines requirements  reporting  response  protection of data  analysis –not to replace or interfere with local response plans Security & Availability Policy Usage Rules Certification Authorities Audit Requirements Incident Response User Registration & VO Management Application Development & Network Admin Guide VO Security

26. Athens, 19 April 2006 26 Operational Security Coordination Team (OSCT) What it is not: –Not focused on middleware security architecture –Not focused on vulnerabilities (see Vulnerabilities Group) Focus on Incident Response Coordination –Assume it’s broken, how do we respond? –Planning and Tracking Focus on ‘Best Practice’ –Advice –Monitoring –Analysis Coordinators for each ROC SSC1 - Job Trace SSC2 - Storage Audit Infrastructure HANDBOOK Incident Response Policy Procedures Resources Reference Playbook Security Service Challenge Infrastructure Agents Deployment Monitoring Tools 3 strategies OSCT membership  ROC security contacts

27. Athens, 19 April 2006 27 Vulnerability Group Purpose: inform developers, operations, site managers of vulnerabilities as they are identified and encourage them to produce fixes or to reduce their impact Set up (private!) database of vulnerabilities –To inform sites and developers Urgent action  OSCT to manage After reaction time (45 days) –Vulnerability and risk analysis given to OSCT to define action – publication? –Will not publish vulnerabilities with no solution Report progress and statistics on vulnerabilities Balance between open responsible public disclosure and creating security issues with precipitous publication

28. Athens, 19 April 2006 28 General management of the User Support process A portal with a well structured information and updated documentation Knowledgeable experts Correct, complete and responsive support Tools to help resolve problems –search engines –monitoring applications –resources status Examples, templates, specific distributions for software of interest Interface with other Grid support systems Connection with developers, deployment, operation teams Assistance during production use of the grid infrastructure

29. Athens, 19 April 2006 29 Central Application (GGUS) Deployment Support Middleware Support Network Support Operations Support TPM ROC 1 ROC 10 ROC… VO Support Interface Webportal The Support Model “Regional Support with Central Coordination" The ROCs, VOs are connected via a central integration platform provided Regional Support units User Support units Technical Support units

30. Athens, 19 April 2006 30 Negotiates resources for user communities Brings together VOs and NGIs Negotiate for services and resources  Run services on behalf of the VO  Provide compute and/or storage resources NOTE: Computational and storage resources are not funded by EGI

31. Athens, 19 April 2006 31 Interaction point with GEANT for grid issues Technical Network Liaison Committee to address grid issues with GEANT/NRENs Definition and establishment of Service Level Agreements for end-to-end services Joint operation of ENOC (e-Infrastructure Network Operations Centre) Deployment of network performance mgmt tools Coordination of input to GGF Network Measurements Working Group

32. Athens, 19 April 2006 32 Foundation Grid Middleware Security model and infrastructure Computing (CE) & Storage Elements (SE) Accounting Information providers and monitoring Applications Higher-Level Grid Services Workload Management Replica Management Visualization Workflows Grid economies etc. Middleware Provide specific solutions for supported applications Host services from many sources More rapid changes than Foundation Grid Middleware Deployed as application software using procedure provided by grid operations Application independent Evaluate/adhere to new stds Emphasis on robustness/stability over new functionality Deployed as a software distribution by grid operations

33. Athens, 19 April 2006 33 Integration VDT/OSG OMII- Europe EGEE Etc. … Testing & Certification Support, analysis, debugging Production service deployment Pre-production service Middleware providers Integration/testing Certification activities Middleware testing and certification feedback

34. Athens, 19 April 2006 34 Software deployment paths Foundation grid middleware and selected high-level services are supported directly within EGI Application and high-level services from other sw providers will be incorporated ProductionPre-production Testing/ Integration Support/ Development Other sw providers componentsreleases components high-level services components