Infrastructure as a Service

1. Infrastructure as a Service Kees Neggers, NSF IRNC Kickoff 13 July 2010

2. Networks: Enablers for progress • The Roman empire: a road system to enable conquest • 18th – 20th centuries: enabling the industrial revolution: • Canals, roads en railroads • Post-Telegraph-Telephone • 20th century: birth of the Digital Economy: • Internet

3. What’s next? ‘New networks’ will remain important enablers for economic and social developments 21th century will need a Cyberinfrastructure (or e-Infrastructure in EU)

4. Trends driving research • System level science • The integration of diverse sources of knowledge about the constituent parts of a complex system with the goal of obtaining an understanding of the system's properties as a whole [Ian Foster] • Inter/trans-disciplinary research • Each discipline can solve only part of a problem • Collaboration between different research groups • Distributed across states, countries, continents • Research driven by (distributed) data • Data explosion, both in volume and complexity • Simulation and experiment combined • Exploring data-sets with no up-front hypothesis • Research carried out using simulation and modeling • HPC and Grid computing together with high-speed networks and data visualizations enable totally new visions in simulations of complex phenomena

5. Modern Research needs an integrated ICT Infrastructure • Providing seamless access to and allowing the shared use of: • Computing and storage facilities • Generic application services • Sensors and instruments • Network resources • Providing hassle free end-to-end connectivity via a single user interface and a single control plane for the allocation of multiple resources, from multiple domains and in multiple locations This can not be provided by a single operator, or even a few operators, but requires worldwide distributed ICT resources connected by advanced networks

6. This is effecting all ICT infrastructure components • Computing • Data • Software • Networking • Organization • Education • At the Internet2 Spring 2010 meeting Alan Blatecky gave a nice introduction of this ongoing evolution titled: “ Cyberinfrastructure Framework for 21st Century Science & Engineering (CF21)”

7. Cyberinfrastructure Ecosystem Organizations Universities, schools Government labs, agencies Research and Medical Centers Libraries, Museums Virtual Organizations Communities Scientific Instruments Large Facilities, MREFCs,telescopes Colliders, shake Tables Sensor Arrays - Ocean, environment, weather, buildings, climate. etc Expertise Research and Scholarship Education Learning and Workforce Development Interoperability and operations Cyberscience Discovery Collaboration Education Data Databases, Data repositories Collections and Libraries Data Access; storage, navigation management, mining tools, curation Computational Resources Supercomputers Clouds, Grids, Clusters Visualization Compute services Data Centers Networking Campus, national, international networks Research and experimental networks End-to-end throughput Cybersecurity Software Applications, middleware Software development and support Cybersecurity: access, authorization, authentication Maintainability, sustainability, and extensibility

8. e-Infrastructure • COM (2009) 108, ICT Infrastructures for e-science: “e-Infrastructure is an environment where research resources (hardware, software and content) can be readily shared and accessed wherever this is necessary to promote better and more effective research” see: http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2009:0108:FIN:EN:PDF • In Europe also the e-Infrastructure Reflection Group recently produced a White Paper 2009 and a Roadmap 2010 addressing e-infrastructure developments, see: http://www.e-irg.eu/

9. What does this all mean • The way research is done is rapidly and radically changing: researchers no longer need access to technologies or products, they need e-infrastructure as a service …But (as Alan said at the I2 meeting, after 4 centuries of constancy) such radical change cannot be adequately addressed with (our current) incremental approach! • Close collaboration among the different providers and users will be essential to create and maintain the required ecosystem • Any monolithic approach is doomed to fail, openness and flexibility should be major design parameters for the architecture

10. NL moves to an integrated e-Infrastructure for Research • SURF will become the single organization to be responsible for the ICT infrastructure • Integrating existing organizations for Research Network, Grid Computing, Supercomputing, and services for e-science • Working towards a single control plane and a consistent set of services for research

11. Building a national knowledge infrastructure http://www.cookreport.com/

12. Modern Research leads to new network demands • Explosion in the amount of data from experiments and simulations; Examples: LHC, LOFAR, e-VLBI Networks are already an integral part of these systems • Need for near real-time processing of very large datasets; Example: LHC Atlas trigger • Increase in remotecollaboration • Distributed sensors • Shared computing and storage, grids • Virtual teams …and many new users will not be ICT experts

13. Network challenges • Today’s Internet is not good enough to support the needed e-Infrastructure for research • Fit for delay tolerant, many-to-many communication • No guaranteed services on a “best effort” network • To support an e-Infrastructure Research networks will have to do better... • Provide guaranteed performance for large data flows and time-critical applications • Support increasingly heterogeneous access methods • Take into account security and environmental issues And above all, must be able to support the most advanced and demanding users

14. Relying on commercial operators will be risky • Operators with a legacy business model will: • Attempt to retain the traditional telephony model • Assume that network resources are scarce • Attempt to move as high as possible in OSI stack to “create value” • Operator driven standardization efforts are based on this model • MPLS (Multiprotocol Label Switching)used to create IP-VPN’s even where lightpaths would be better • UMA (Unlicensed Mobile Access) attempts to integrate WiFi in cellular business model • IMS (IP Multimedia Subsystem) tries to put the operator in charge again

15. And in addition… ‘It is uncertain who will invest in new generations of infrastructure and what model will prevail.’ From: Trends in connectivity technologies and their socioeconomic impacts Final report of the study: Policy Options for the Ubiquitous Internet Society Prepared for DG Information Society by the RAND Corporation in 2009

16. Impact of Cloud computing • Clouds already offer advanced services to inexperienced users via easy to use graphical user interfaces • New students arrive with this experience and expects nothing less at the Universities • Researchers used to develop their own ICT tools, they now expects them to be available in the same way they use clouds Users expect no less than an integrated, sustainable and extensible ICT infrastructure

17. Application andContent Infrastructure • Over the years the Internet already evolved from a pure communications infrastructure towards a distributed application and content infrastructure • The commercial application and content providers are already building their own worldwide networks and connect at neutral exchange points to local access providers • The research community also will need a dedicated on-demand global infrastructure, based on hybrid networks and open exchanges NSF-OCI and IRNC have been a major facilitator in the development of open exchanges

18. Towards an open Cyberinfrastructure • Open exchanges have proven to be enablers both for networking and for application development • Next step now is to make all infrastructure resources available on demand via smart and easy to use middleware • Cyberinfrastructure will evolve rapidly over time for many years to follow, hence innovation power will be crucial. To develop new services in time multiple efforts are needed, both complementary and competing, in a globally coordinated effort IRNC could play a leading role

19. We urgently need to … • Update the (now over 15 year old) governance, organizational and financial structure of research networking to allow us to: • Stay ahead of commercial operators • Better involve users, both in planning and reviewing the services • Integrate network service with other ICT providers • Create the needed diversity in the ecosystem (monocultures will die) • Invest in open exchanges • Secure funding Think global, act local

20. NetherLight

21. Thank you for your attention SURFnet: Engine for Innovation