Presentation on theme: "Software Defined Networking"— Presentation transcript:
1Software Defined Networking Nick McKeownStanford University
2(part 1)My talk will be in three parts. I want to keep you on your toes, so I’m not going to tell you what the 2nd and 3rd parts are yet.To get the ball rolling, I thought I’d start by telling you a little bit about what I do.
3Why I love my job I work with people much smarter than me. I get to work on intellectually interesting ideas.… that might positively change the practice.Then, we try to actually change the practice.
4Whatever it takes Prove a theorem Write a paper Build a demo Talk to lots of industry peopleWrite a standardGive lots of talksWrite a blogStart a companyBuild an open source tool…
5Choosing research projects Pick a problem that is intellectually interesting.And improves the practice.And industry doesn’t like (yet).
6Ethane Martin Casado et al [Sigcomm ‘07] Policy“Laptops can’t accept incoming connections”“A can’t talk to B”Network Control PlaneControlPacketForwardingPacketForwardingPacketForwardingPacketForwardingPacketForwarding
7Microsoft: “Come on in….” Cisco: “It will never work…”Raw nerve.We must be onto something.
8Checklist Intellectually interesting. And improves the practice. And industry doesn’t like (yet).
9Corollary: You can’t give stuff away Example 1: Order of magnitude faster router (1997)Tried to give Tiny-Tera away for free.Industry wasn’t ready.Started Abrizio.Example 2: Network Memory (2001)Tried to give it away for free, to save $500M per year.Started Nemo.Example 3: Ethane (2007)Tried to give it away for free; early stages of SDN.Started Nicira.
10I put everything in public domain IndustryInvests huge amounts to develop and sell products.Patents protect ideas, giving the confidence to invest.University researchServes society at large.Stay ahead by running fast, not by protecting.Makes it easier to work with industry.If it’s good research, industry doesn’t see it yet.Everything in public domain since 1999.
11(part 2)In part 2 of my talk I’m going to tell you about software defined networking. What it is, what it isn’t. What are the technical consequences and what will be some of the business consequences.
12Urs Hölzle (Google), ONS 2012 If you are in any doubt about whether OpenFlow/SDN will be deployed in the WANUrs Hölzle (Google), ONS 2012
14(when we clear away all the hype) What is SDN?(when we clear away all the hype)
15SDN is the separation of the control plane from the forwarding plane.
16Software Defined Network (SDN) ControlProgramControlProgramControlProgramGlobal Network MapNetwork OSPacketForwardingAbstractForwardingModel(e.g. OpenFlow)PacketForwardingPacketForwardingPacketForwardingPacketForwarding
17The Technical Benefits (1) Well-defined control abstractionControl plane can run on modern serversCan adopt software engineering best-practicesEasier to add new control programs…or customize locallySolve distributed systems problem once, rather than for every protocol
18OSPF OSPF Dijkstra Network OS OS Dijkstra 5% 95% Specialized Hardware PacketForwardingGlobal Network MapNetworkMap95%OSSpecializedHardware
22OpenFlow Philosophy Long-term, forwarding looking Match: Very general, not protocol specific.Action: Small instruction set, not protocol specific.Make it easy to add new headers and actions.Any network (packet, circuit, radio).Short-term, backward lookingMatch: include well-known header fields.Action: necessary set for existing protocols.Support existing protocols on existing switch chips.
24The Technical Benefits (3) Well-defined forwarding behaviorThe forwarding tables capture the entire forwarding behavior.Control plane writes the forwarding state.Therefore, we can verify its correctness.
27Software Defined Network (SDN) “A can talk to B”“Guests can’t reach PatientRecords”“No loops”PolicyControlProgramControlProgramControlProgramGlobal Network MapNetwork OSMatchActionAAction(A)BAction(B)CAction(C)MatchActionAAction(A)GAction(G)HAction(H)PacketForwardingPacketForwardingMatchActionXAction(X)YAction(Y)ZAction(Z)ForwardingBehaviorMatchActionFAction(F)GAction(G)HAction(H)PacketForwardingPacketForwardingMatchActionAAction(A)GAction(G)DAction(D)PacketForwarding
28Networks notoriously hard to debug Today, even simple questions hard to answer:Can host A talk to host B?What are all the packet headers from A that can reach B?Are there any loops in the network?Is Group X provably isolated from Group Y?What happens if I remove a line in the config file?
29Header Space Analysis A B Header Data Header Data L 01110011…1
33Header Space Analysis [Kazemian NSDI ‘12] ConsequencesAbstract forwarding model; protocol independentFinds all packets from A that can reach BFind loops, regardless of protocol or layerCan prove that two groups are isolatedCan verify if network adheres to policy
34HSA as a “foundation” Analogy to Boolean algebra for logic design HSA enables many tools and methodsIndependent static checkingIn-line in-controller invariance checkingDynamic testing: Automatic test packet generationDynamic testing: Automatic performance monitoringAnalogy to Boolean algebra for logic design
36The Business Consequences Vertical integration will finally be replaced by a more competitive industry with a level playing field.Large growth in software industry for networking. Faster innovation.Hardware switches will focus on capacity, fan-out and power. (Think Intel for networking).Multi $Bn tools industry will emerge.
37Corollary The customer and end-user will be better served Networks will be cheaper.Networks will be faster.Networks will be (much) more reliable.Networks will improve faster.Networks will be in service of the owner, the operator, the customer and the application rather than just the high-margin vendor.SDN is in the best interest of society at large.To resist is to be protectionist and self-serving.
40Tens of millions of lines of code. Closed, proprietary, outdated. SpecializedFeaturesHundreds of protocols6,500 RFCsSpecializedControlPlaneTens of millions of lines of code.Closed, proprietary, outdated.SpecializedHardwareBillions of gates.Power hungry and bloated.
41“My box now has an OpenFlow interface too!” What SDN isn’tRam in even more lines of code…“My box now has an OpenFlow interface too!”
43What’s new? Separation of control from forwarding. Programmatic control of forwarding by writing entries into tables.Transport networks have done this for decades!
44Natural Evolution: Converged Control ProgramControlProgramControlProgramGlobal Network MapNetwork OSDirect Control ofHardware, orVirtual transport networkControlData CenterData CenterTransportNetworkMPLSMPLSNew ONF Working Group: Optical Transport
47Why it was supposed to happen Technology: High capacity optical crossconnects. Aggregation: High capacity packet networks. Cost: Circuit switches cost less; use less power. Standard: GMPLS.
48Why GMPLS failed Router vendors prevented it. GMPLS standard much too complicated.
49OSPF-TE, RSVP-TE + many more Packet NetworkTransport NetworkIP/MPLS Control PlaneOSPF-TE, RSVP-TE + many moreGMPLS Control PlaneOSPF-TE, RSVP-TEEMSEMSEMSProprietary InterfaceUNIVendor IslandsWe Didn’t Make it Easy!49
50OSPF-TE, RSVP-TE + many more Packet NetworkTransport NetworkIP/MPLS Control PlaneOSPF-TE, RSVP-TE + many moreGMPLS Control PlaneOSPF-TE, RSVP-TEOSPF-TE, RSVP-TEOSPF-TE, RSVP-TEEMSEMSEMSProprietary InterfaceUNIVendor IslandsGMPLSOSPFRSVP175kQuagga50kLinux50
51Good Architecture Simplifies 3% as much code!AggregationDynamic BandwidthRecovery4.7kGlobal Network Map68kNetwork OSNOXLinuxQuaggaLinuxOSPFRSVP50k175kGMPLSPacketNetworkPacketNetworkTransportNetwork
52ConclusionsSDN is here to stay: It introduces the right abstractions into network control.It will sweep away protectionist practices, and level the playing field.It will enable converged control of packet and transport networks.I don’t recommend sitting on the sidelines.