Carmelita Görg Universität Bremen
Transcription
Carmelita Görg Universität Bremen
Looking 4ward t the to th Future F t I Internet t t Carmelita Görg Universität Bremen Contents Introduction – who, why and what? Migration g and Innovation – 1000 networks bloom How can we build these networks? What do we want or connect in these networks – people, things, information ... How to transport data? What is a path? How can we manage these networks? A Summary 23/11/2008 PM/Slide 2 © 4WARD Consortium Confidential Geographical distribution Ericsson Nokia-Siemens-Networks VTT Ericsson KTH SICS Alcatel-Lucent Deutsche Telecom Ericsson Nokia-Siemens-Networks TU Berlin Univ. of Bremen Univ. of Karlsruhe UNiv. of Paderborn Canada NEC Univ. of Lancaster Univ. off Surrey S Ericsson Finland Sweden Norway Telekomunikacja Polska WIT Ireland UK Alcatel-Lucent France Telecom GET INT GET-INT LIP6 Siemens TPUCN Poland Germany Austria France Romania Switzerland Italia IST-TUL PTIN Spain Israel Univ. of Basel Telcom Italia US Rutgers University Technion Robotiker-Tecnalia Telefonica © 4WARD Consortium Acknowledgements 4WARD Consortium 4WARD is i funded f d d by b the th European E C Commission i i under the EU‘s 7th Framework Programme www.4ward-project.eu www 4ward project eu Norbert Niebert, The Way 4WARD to the Creation of a Future Internet, ICT Mobile Summit 2008, Stockholm, June 2008 CEWIT 2008 Network (R)Evolution – How? Is I it just j t IPv7 IP 7 or IPv8 IP 8 … or creating ti an alternative? lt ti ? By adding and patching we run into the risk of “obesity obesity and a weak immune system” system Or should we even dare to think of tailor-made networks, fit for the purpose and reliable? Will migration work? How can we solve this? 23/11/2008 PM/Slide 5 © 4WARD Consortium Confidential The Facets of 4WARD Business Innovation Architecture Framework © 4WARD Consortium Fo olding Point Endpoint Forwarder Combination of clean slate clean-slate research approaches pp to address the Network of the Future Size: Roughly 23 M€ Time 2 years Network Virtualisation – Main Innovations Network virtualisation as a meta meta-architecture architecture in a commercial setting – Enable co-existence of diverse network architectures – Enable deployment of innovative approaches – Enable new business roles and players • Allow split of infrastructure-/network-/service-providers • Lower barriers of entry • „Market place“ for shareable network resources Provisioning and virtualisation management framework – On-demand instantiation of virtual networks at large scale Virtualisation of diverse resources in a common framework – Routers, links, servers – can all be done today but need a unifying e2e approach – Extension on the virtualisation of the wireless infrastructure and spectrum – Folding points providing interworking between virtual networks 23/11/2008 PM/Slide 7 © 4WARD Consortium Confidential ee Folie 7 eednni1 Need a slide on folding points. The rest is doable. Too much text on the slide - I deleted a few subb ullets. Norbert Niebert; 17.09.2008 Possible Provider Roles in a Virtualised Environment End-to-End Deployment y Virtual Network Operator VNet VNet VNet VNet Substrate VNet VNet Phys. Net Infrastructure Broker (optional) Substrate Legacy Operator Substrate Vertically Integrated Operators (virtualised networks) 23/11/2008 Substrate Substrate Infrastructure Providers PM/Slide 8 © 4WARD Consortium Confidential Virtualised Networks Virtualisation of diverse resources in a common framework Enable co‐existence of diverse network architectures Dynamic management of virtual networks Enable new business roles and players Folding Points V t Operator Vnet O t Domain D i Vnet Provider Domain Vnet‐Provider A Inf‐Provider A Vnet‐Provider B Routers Servers Routers, Servers, ... Infrastructure Provider Domain Links, ... Wireless, spectrum, ... Inf‐Provider B 23/11/2008 Inf‐Provider C PM/Slide 9 © 4WARD Consortium Confidential Virtual Radio Framework Slice 1 Slice 2 Slice 3 Slice 4 c-plane u-plane c-plane u-plane c-plane u-plane c-plane u-plane Control Slice Mgr. HL Control HL Control HL Control HL L2 L2 L2 L2 PHY PHY PHY PHY RAC (per ( slice li resource block bl k allocation) ll ti ) harmonised resource access Virtualisation of Wireless Resources & Efficient Spectrum Sharing Flexible and cost-efficient d l deployment t off new radio di technologies Harmonised access of slices to a common radio resource block Slices can implement their own protocols/methods – routing, mobility management, g naming – radio protocols, channel coding, smart antenna steering – cross layer optimisation Scheduling and isolation 23/11/2008 PM/Slide 10 © 4WARD Consortium Confidential Virtual Router Performance • • High-performance virtual router platform for modern commodity hardware Evaluation of virtualised forwarding planes in terms of isolation and fairness 23/11/2008 PM/Slide 11 © 4WARD Consortium Confidential VNet Instantiation Process VNet Provider Install & Configure Query Offer (abstracted network descr. (specific network descr. Request + Grant + (Specific(access networkIDs description) query language) + cost and other parameters) certificates) Initiate Inter-Provider Links Commit Mapping + + Embed Commit-Request VMGW Peering Point Infrastructure Provider A VMGW Commit + Embed VMGW Peering Point Infrastructure Provider C 23/11/2008 Commit + Embed Infrastructure Provider B PM/Slide 12 © 4WARD Consortium Confidential How far have we come? Query Language VN requirements • • Queue VN Request analysis and processing VN Provider Scheduling VN Provisioning g and Management g Architecture Security Management VN Mapping Framework (VN mapping algorithms and protocols) - Authorisation - Authentication Accounting Mobility Mgt, QoS Mgt…. Resource description (Language/Semantic/ Identifiers/Names…) Resource Monitoring & Discovery Distributed Database Resource Information Database Update Configuration Action Agents Virtual Resources (virtual nodes/links, CPU, memory, bandwidth…) Virtualisation Management WP3-Task3.2 Resource Virtualisation Abstract view of substrate from VN provisioning & management point of view Virtual Node Virtual Networks Probes/monitors Substrate Path Substrate Node Virtual Link Substrate Link Substrate Network PM/Slide 13 © 4WARD Consortium Confidential Resource Allocator & configurator Physical Reso ources Layer 23/11/2008 VN Requests Virtual Resou urces Layer • • • VN customers Provisioning & Manageme ent Layer • • Draft D ft architecture hit t Scalable mapping algorithms using data mining technology Initial definition of signaling and control interfaces First version resource description language – Modelling of resources and networks – XML-based – Used for request and offer – Additional query language for complex requests Virtual Radio concept p Early prototyping and testbeds Controlled Interworking concept Language/Semantic/Ide/Syntax e.g.: topolgy description and objectives But how to design a network architecture? By B reusing i and d patching t hi existing i ti protocols t l we forgot f t to t develop tools for clean slate design esp. for the more detailed network architecture specifications Can such a design toolkit and process be easily developed and used? Where can we find reusable components other than protocol specs and implementations? How can we ensure interoperability? p y 23/11/2008 PM/Slide 14 © 4WARD Consortium Confidential Overview of 4WARD’s Network Design Process Requirements … (Meta) Modeling Model Driven Design Process (including iterations) Network architect (needs to have knowledge about networks – works off-line, before (!) operation of the network) Composition Building blocks, of functionalityy Netlets Netlets, (CFI) Abstract strata, Architecture patterns Interoperability “Blue Print” of network Architecture (selected netlets/strata) Netlets, Strata Repository 4WARD Architectural Framework 23/11/2008 Prototype Repository I l Implementation t ti PM/Slide 15 © 4WARD Consortium Confidential What a new network architecture could interconnect We W are used d to t think thi k a network t k consisting of nodes (end + forwarding) and links What if we start to network the information we are looking for? Triggered by Van Jacobsen and others a new view on interconnecting i f information ti has h emerged d ... that could change the way we engineer networks fundamentally The Networking of Information is looking into this from a systems y perspective p p 23/11/2008 PM/Slide 16 © 4WARD Consortium Confidential Information Objects and Data Objects Aggregation IOs Encoding IO Data Obj t Objects Files All About Song1 Song1 Paris Virtual Entity IO Service IO Eiffel Song1 mp3 Song1.mp3 Song1 wav Song1.wav Lyrics txt Lyrics.txt Eiffel jpg Eiffel.jpg Song1.mp3 S 1 3 Song1.mp3 Song1.wav S 1 Song1.wav Lyrics.txt Lyrics.txt Lyrics.txt Eiffel.jpg ff Eiffel.jpg 23/11/2008 Service1 Service1 Service1 PM/Slide 17 © 4WARD Consortium Confidential How to transport? Can C we assume that th t turning t i wireless i l and d optical ti l media di into i t copper will work forever? And why is it efficient to do transport innovations only as overlays? What can be gained with a completely fresh view on transport mechanisms? The Generic Path is an answer to these questions... 23/11/2008 PM/Slide 18 © 4WARD Consortium Confidential An architecture and a set of mechanisms the th Generic G i Path P th architecture hit t – a much richer class of data flows, beyond TCP, UDP – state within the network, as necessary y but no more than necessaryy – common management interfaces, to set up and tear down flows and to query their status – explicit p identification,, notably y to facilitate control of multi-flow applications like videoconferencing mechanisms for assured performance and efficient operation – to exploit techniques like network coding and cooperative transmission – to choose the "best" paths for the considered transport – to ensure resource sharing is "fair" fair and meets application requirements – to manage the mobility of users, networks and information 23/11/2008 PM/Slide 19 © 4WARD Consortium Confidential Management?! The Th mostt urgentt need d in i a dynamic d i world ld is i Self-Management Automation of Management has been a research topic for many years Does it provide in practice more than automated settings on FI routers? Can we rely on this? What are the new approaches in this area? 23/11/2008 PM/Slide 20 © 4WARD Consortium Confidential Monitoring and adaptation • INM ffunctions i aim i at predictability 1 fraction o of aggregating nodes h=2 – Local optimization loops reduce time to react – Trade-off between accuracy, timeliness and overhead – Tunable objectives in adaptation algorithms – Anomaly detection to perform isolate exceptions in the network h=1 0,8 h=3 h=4 0,6 0,4 0,2 0 1 10 100 Updates/sec 1000 Objectives NetInf • Final objective – Build full control control loops – Maintain service-level objectives – Enforcing required OAM functions VNet GP 23/11/2008 PM/Slide 21 © 4WARD Consortium Confidential Summary 4WARD follows f ll a number b off technical t h i l innovation i ti approaches in all areas of future networking They all start from a clean clean-slate slate perspective, perspective defining radically new solutions for the Network of the Future After 11 month work first results look promising Now the hard work of refining and integrating starts towards a new Network of the Future as a Family of Networks 23/11/2008 PM/Slide 22 © 4WARD Consortium Confidential FISS 09 FISS´09 Future F t Internet I t t Summer S School S h l 2009 Universität Bremen July J l 20 20-24, 24 2009 Mixture of courses, presentations and invited talks by 4ward in cooperation with other Future Internet projects projects, e e.g. g EuroNF, ANA, Chianti, … Courses for g graduate students and researchers See 4ward website or www.comnets.uni-bremen.de ComNets FFV 21. Nov. 08 Slide 23 © 4WARD References Ines Houidi, Wajdi Louati and Djamal Zeghlache, “A Distributed Virtual Network Mapping Algorithm,” In Proceedings of the 2008 IEEE International Conference on Communications (ICC 2008), May 19-23, 2008 Beijing 2008, Beijing, China China, pp pp. 5634 – 5640 Ines Houidi, Wajdi Louati and Djamal Zeghlache, “A Distributed and Autonomic Virtual Network Mapping Framework,” In Proceedings of the 4th International Conference on Autonomic and Autonomous Systems y (ICAS ( 2008), IEEE Computer Society, March 16-21, 2008, Gosier, Guadeloupe, pp. 241-247 N. Egi, A. Greenhalgh, M. Handley, M. Hoerdt, F. Huici, and L. Mathy; “Fairness Fairness Issues in Software Virtual Routers” Routers , in Proc Proc. PRESTO '08 08, ACM SIGCOMM Workshops, August 2008, Seattle, USA. Joachim Sachs, Stephan Baucke, “Virtual Radio – A Framework for Configurable Radio Networks”, The Fourth International Wireless Internet C f Conference (WICON 2008), 2008) November N b 2008, 2008 M Maui, i USA 23/11/2008 PM/Slide 24 © 4WARD Consortium Confidential