GENI: Global Environment for Network Innovations. Larry Peterson, et al. Kideok Cho

Transcription

1 GENI: Global Environment for Network Innovations Larry Peterson, et al. Kideok Cho

2 -2/37- Three steps to reach Future Internet Develop component technologies Antenna, transmission, MAC, QoS. Put them into one architecture FIND Experiment in large scale, identify problems PlanetLab, GENI, VINI.. From course overview slide

3 -3/37- Outline What is GENI? Need for GENI GENI Design Physical network substrate Global management framework Summary

4 -4/37- What is GENI? Global Environment for Network Innovations An experimental facility intended to enable fundamental innovations in networking and distributed systems rapidly and effectively embed within itself a broad range of experimental networks interconnec these experimental networks as appropriate with other experiments and the existing Internet provide these networks with users and an operating environment, and rigorously observe, measure, and record the resulting experimental outcomes.

5 -5/37- Need for an Experimental Facility Goal: Seamless conception-to-deployment process Deployment Analysis Simulation / Emulation Experiment At Scale (models) (code) (results) (measurements)

6 -6/37- Chasm Maturity Global Experimental Facility Small Scale Testbeds Deployed Future Internet This chasm is a major barrier to realizing the Future Internet Simulation and Research Prototypes Foundational Research Time

7 -7/37- FAQ1 How is GENI different from a traditional network testbed? GENI will be a general-purpose facility essentially no limits on the network architectures, services, and applications that can be evaluated allows both clean-slate designs and experimentation with real users under real-world conditions

8 PlanetLab New paradigm for network research infrastructure Support experimental validation of new services simultaneously support real users and clean slate designs Key ideas Slice Substrate resources bound to a particular experiment Virtualization multiple architectures on a shared infrastructure Programmable virtually no limit on new designs -8/37-

9 -9/37- PlanetLab (cont d) 645 machines spanning 310 sites and 35 countries nodes within a LAN-hop of > 3M users Supports distributed virtualization each of 375 network services running in their own slice Carries real user traffic 3-4 TB per day

10 -10/37- Slices

11 -11/37- Slices

12 GENI Design

13 -13/37- GENI Requirements Service/architecture neutrality Edge diversity Ease of user access Global reach Instrumentation and data analysis Federation and sustainability Inter-slice composition Policy and governance

14 GENI Design Key Idea Slices embedded in a substrate of networking resources Two central pieces Physical network substrate expandable collection of building block components nodes / links / subnets Global management framework knits building blocks together into a coherent facility embeds slices in the physical substrate -14/37-

15 -15/37- FAQ2 What are the main technical ideas exploited by GENI? the substrate components will be programmable the substrate will be virtualizable allow end-users to seamlessly opt-in to experimental services GENI will be modular, with a well-defined architecture and set of interfaces

16 -16/37- National Fiber Facility Current configuration of the National Lambda Rail (NLR), with includes 26 PoPs across the U.S.

17 -17/37- + Customizable Routers

18 -18/37- + Clusters at Edge Sites Tail circuit

19 -19/37- + Wireless Subnets

20 -20/37- + ISP Peers Legacy Internet Legacy Internet

21 -21/37- Closer Look backbone wavelength Sensor Network backbone switch Dynamic Configurable Switch Customizable Router Internet Wireless Subnet Edge Site

22 -22/37- Summary of Substrate Node Components edge devices customizable routers optical switches Bandwidth (link capacity) national fiber facility tail circuits (including tunnels) Wireless Subnets urban based Mesh wide-area 3G/WiMax cognitive radio sensor net emulation

23 -23/37- Management Framework Management Services GMC - set of interfaces ( plug in new components) - support for federation ( plug in new partners) Substrate Components

24 -24/37- Hardware Components Substrate HW Substrate HW Substrate HW

25 -25/37- Virtualization Software Virtualization SW Virtualization SW Virtualization SW Substrate HW Substrate HW Substrate HW

26 -26/37- Component Manager Control Interface Sensor Interface CM Virtualization SW Substrate HW CM Virtualization SW Substrate HW CM Virtualization SW Substrate HW

27 -27/37- GENI Management Core (GMC) Resource Controller Slice Manager Auditing Archive GMC slice_spec node control sensor data CM Virtualization SW Substrate HW CM Virtualization SW Substrate HW CM Virtualization SW Substrate HW

28 -28/37- GMC (Cont d) Slice Manager records the state of each slices Slice spec, set of components, contact information for the researcher Used to create and control slices Resource Controller records State, location, and capability of each resource What software runs on the resource Auditing archive Periodically uploads and archives the audit data

29 -29/37- Federation GMC GMC...

30 -30/37- User Front-End(s) GUI Front-End (set of management services) provisioning service file & naming service information plane GMC GMC...

31 -31/37- Distributed Services Goals Complete the GENI management story Lower the barrier-to-entry for researchers (students) Example focus areas Provisioning Information Plane Resource Broker Development Tools Security Topology Discovery File & Naming Legacy Internet

32 -32/37- GENI Requirements (revisited) Service/architecture neutrality Edge diversity Ease of user access Global reach Instrumentation and data analysis Federation and sustainability Inter-slice composition Policy and governance

33 -33/37- FAQ3 The deployed Internet represents an enormous investment. How is a clean slate design possibly going to replace it? GENI is designed to enable clean slate thinking that is not constrained by today's Internet but this does not imply that the outcome will necessarily be an entirely new Internet. In other words, "clean slate" is a process, not a result

34 -34/37- FAQ4 What is the relationship between GENI and PlanetLab? PlanetLab serves as a prototype of GENI, and helps make the case that such a facility is feasible However, GENI goes far beyond PlanetLab in several dimensions GENI will also support a richer set of node technologies and expose low-level network (link) behavior, thereby enabling experiments "inside" the network GENI provide a richer set of support services, lowering the barrier for researchers to be able to take advantage of GENI

35 -35/37- FAQ5 What is FIND and how does it relate to GENI? Future Internet Design is an architecturally focused research effort that would benefit from the GENI facility

36 -36/37- Testbeds in Korea KOREN2, Kreonet TEIN2 BcN testbeds PlanetLab IPv6 (Koreav6) Focused on experimental network testbeds Small and medium scale Separate testbeds No shared, federated testbeds From course overview slide

37 -37/37- Summary GENI is an open, large-scale, realistic experimental facility that will revolutionize research in global communication networks. A central goal of GENI is to change the nature of networked and distributed systems design; to move from a fully empirical to a more rigorously understood design process.

38 Appendix

39 -39/37- GENI customizable high-speed router

40 -40/37- Optical Fiber ROADM: Reconfigurable Optical Add/Drop Multiplexer

41 -41/37- Planning Group Tom Anderson, Washington Dan Blumenthal, UCSB Dean Casey, Ngenet Research David Clark, MIT Deborah Estrin, UCLA Larry Peterson, Princeton (Chair) Dipankar Raychaudhuri, Rutgers Mike Reiter, CMU Jennifer Rexford, Princeton Scott Shenker, Berkeley John Wroclawski, USC/ISI

42 -42/37- Motivating Research Challenges (1/2) Limits of today s Internet It is not secure It cannot deliver to society the potential of emerging technologies It does not provide adequate levels of availability It creates barriers to economic investment and enhancement by the private sector It was not designed to make it easy to set up, to identify failures and problems, or to manage

43 -43/37- Motivating Research Challenges (2/2) Future Internet must enable and encourage a world: mobility and universal connectivity is the norm more and more of the world s information is available online smarter by the effective use of sensors and controllers balanced realization of important social concerns computing and networking is no longer something we do, but a natural part of our everyday world

44 Existing Tools Simulators ns2 Emulators Emulab WAIL Wireless Testbeds ORBIT Emulab Wide-Area Testbeds PlanetLab RON X-bone DETER -44/37-

45 Today s s Tools Have Limitations Simulation based on simple models topologies, admin policies, workloads, failures Emulation (and in lab tests) are similarly limited only as good as the models Traditional testbeds are targeted often of limited reach often with limited programmability Testbed dilemma production: real users but incremental change research: radical change but no real users -45/37-

46 Project Management Structure

47 Project Management GENI Community Consortium (GCC) Executive Committee (EC) provides project oversight appoints Project Director and Project Manager Technical Advisory Board (TAB) provides technical leadership for the project establishes a set of working groups Project Management Office (PMO) includes financial, operations and planning, workflow, system engineering office -47/37-

48 -48/37- Project Management (cont) Executive Committee Project Director Chief Architect Project Manager Technical Advisory Board Research Coordination WG Project Management Office Finance Office Facilities Architecture WG Backbone Network WG Wireless Subnet WG Designs Systems Engineering Office (Chief Engineer) Legal Affairs Office Operations & Planning Office External Liaison Office Distributed Services WG Education & Outreach WG Specifications Contracting & Supervision Education & Outreach Office Contracted Development & Assembly Teams

49 -49/37- Working Groups Research coordination Coordinate use of the facility Clark (co-chair), Shenker (co-chair) Architecture define core modules & interfaces Peterson (co-chair), Wroclawski (co-chair), Barford, Brassil, Schwab Backbone fiber facility / routers & switches / tail circuits / peering Rexford (chair), Blumenthal, Casey, Lakshman, McKeown, Turner, Zhang

50 -50/37- Working Groups (cont) Wireless RF technologies / deployment Raychaudhuri (chair), Bahl, Estrin, Evans, Gerla, Heidemann, Minden, Seshan Services edge sites / infrastructure & underlay services Anderson (chair), Andersen, Kaashoek, Pai, Reiter, Roscoe, Stoica, Vahdat Education training / outreach / course development TBD