1. OpenFlow/Software Defined Networks 1

2. Exec Summary OpenFlow/SDN enables innovations within – Enterprise, backbone, & data center networks – Represents a promising architecture direction Providers like it for their own reasons – Enabling an ecosystem OpenFlow/SDN networks are being deployed – R&E networks around the world Time for regionals to get involved. Take advantage of GPO Solicitation 3.

3. Internet has many problems Plenty of evidence and documentation Internet’s “root cause problem” It is Closed for Innovations 3

4. Million of lines of source code 5400 RFCsBarrier to entry Billions of gates BloatedPower Hungry Many complex functions baked into the infrastructure OSPF, BGP, multicast, differentiated services, Traffic Engineering, NAT, firewalls, MPLS, redundant layers, … An industry with a “mainframe-mentality”, reluctant to change The Ossified Network Specialized Packet Forwarding Hardware Operating System Feature Routing, management, mobility management, access control, VPNs, … 4

5. The SDN Approach Separate control from the datapath – i.e. separate policy from mechanism Datapath: Define minimal network instruction set – A set of “plumbling primitives” – A vendor-agnostic interface: OpenFlow Control: Define a network-wide OS – An API that others can develop on 5

6. Specialized Packet Forwarding Hardware Feature Specialized Packet Forwarding Hardware Operating System Operating System Operating System Operating System Operating System Network OS Feature Restructured Network 6

7. Feature Network OS 1. Open interface to hardware 3. Well-defined open API 2. At least one Network OS probably many. Open- and closed-source The “Software-defined Network” OpenFlow 7 Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware

8. OpenFlow Basics Narrow, vendor-agnostic interface to control switches, routers, APs, basestations. 8

9. Network OS Step 1: Separate Control from Datapath 9 OpenFlow Switch OpenFlow Switch OpenFlow Switch OpenFlow Switch

10. Step 2: Cache flow decisions in datapath “If header = x, send to port 4” “If header = ?, send to me” “If header = y, overwrite header with z, send to ports 5,6” 10 OpenFlow Switch OpenFlow Switch OpenFlow Switch OpenFlow Switch Flow Table Flow Table

11. Plumbing Primitives 1.Match arbitrary bits in headers: – Match on any header; or new header – Allows any flow granularity 2.Actions: – Forward to port(s), drop, send to controller – Overwrite header with mask, push or pop – Forward at specific bit-rate 11 Header Data Match: 1000x01xx0101001x

12. Feature Network OS 1. Open interface to hardware 3. Well-defined open API 2. At least one Network OS probably many. Open- and closed-source The “Software-defined Network” OpenFlow 12 Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware

13. Network Operating System 1 Open interface to hardware Virtualization or “Slicing” Layer (FlowVisor) Network Operating System 2 Network Operating System 3 Network Operating System 4 Feature Many operating systems, or many versions Open interface to hardware Isolated “slices” Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Simple Packet Forwarding Hardware Feature

14. Some research examples at Stanford 14

15. FlowVisor Creates Virtual Networks OpenFlow Protocol FlowVisor OpenPipes Demo OpenFlow Wireless Demo OpenFlow Protocol PlugNServe Load-balancer OpenPipes Policy Multiple, isolated slices in the same physical network OpenFlow Switch OpenFlow Switch OpenFlow Switch [Sigcomm 2009 – Best Demo] [Paper in submission]

16. Demo Infrastructure with Slicing

17. OpenPipes Partition hardware designs across a network 17 [Sigcomm 2009 – 2 nd Best Demo] [Paper in submission]

18. Load-balancing as Network Primitive 18 OpenFlow Switch OpenFlow Switch OpenFlow Switch OpenFlow Switch Internet OpenFlow Switch [Sigcomm 2009 Demo] [Paper in preparation] Goal: Minimize http response time over campus network Approach: Route over path to jointly minimize Network OS Load- Balancer “Pick path & server”

19. Intercontinental VM Migration Moved a VM from Stanford to Japan without changing its IP. VM hosted a video game server with active network connections. [Sigcomm 2008– Best Demo]

20. Feature NOX Converging Packet and Circuit Networks IP Router IP Router TDM Switch TDM Switch WDM Switch WDM Switch WDM Switch WDM Switch IP Router IP Router Goal: Common control plane for “Layer 3” and “Layer 1” networks Approach: Add OpenFlow to all switches; use common network OS OpenFlow Protocol OpenFlow Protocol [Supercomputing 2009 Demo] [OFC 2010]

21. ElasticTree Goal: Reduce energy in data center networks Approach: 1.Reroute traffic 2.Shut off links and switches to reduce power [NSDI 2010] Network OS DC Manager DC Manager “Pick paths”

22. ElasticTree Goal: Reduce energy in data center networks Approach: 1.Reroute traffic 2.Shut off links and switches to reduce power [NSDI 2010]XX X X X Network OS DC Manager DC Manager “Pick paths”

23. Exec Summary OpenFlow/SDN enables innovations within – Enterprise, backbone, & data center networks – Represents a promising architecture direction Providers like it for their own reasons – Enabling an ecosystem OpenFlow/SDN networks are being deployed – R&E networks around the world Time for regionals to get involved. Take advantage of GPO Solicitation 3.

24. 24 New Data Center Cost 200,000 servers Fanout of 20  10,000 switches $5k commercial switch  $50M $1k custom-built switch  $10M Savings in 10 data centers = $400M Control Optimize for features needed Customize for services & apps Quickly improve and innovate The value prop applies to enterprise and service provider networks

25. Cellular Networks Recently made transition to IP Billions of mobile users Need to securely extract payments and hold users accountable IP is dreadful at both, yet hard to change 25

26. Telco Operators e.g. AT&T, DT, NTT, … Global IP traffic will grow 5x by 2013 End-customer monthly bill remains unchanged Therefore, CAPEX and OPEX need to be reduced 5x by 2013 But in practice, reduces by <20% per year Q: How can operators reduce cost? Q: How can they differentiate their service? A: SDN is a potential solution they want to explore 26

27. Now, introducing: Srini Seetharaman seethara@stanford.eduseethara@stanford.edu

28. Matt Davy’s Strawman Approach to a Regional Deployment of OpenFlow

29. OpenFlow building blocks Controller Nicira NOX Slicing Software FlowVisor Console 29 Applications LAVI ENVI (GUI) GENI Aggregate Manager GENI Aggregate Manager n-Casting NetFPGA Software Ref. Switch Software Ref. Switch Broadcom Ref. Switch Broadcom Ref. Switch OpenWRT PCEngine WiFi AP Commercial Switches Stanford Provided OpenFlow Switches SNAC Stanford Provided Monitoring/ debugging tools oflops oftrace openseer OpenVSwitch HP, NEC, Pronto, Juniper.. and many more

30. OpenFlow Vendor Hardware more to follow... NEC IP8800 HP ProCurve 5400 and others Juniper MX-series Cisco Catalyst 6k (prototype) Core Router Enterprise Campus Data Center Circuit Switch Wireless Pronto Prototype Product Ciena CoreDirector WiMAX (NEC) Arista 7100 series (Q4 2010) 30

31. Commercial Vendors ModelPrice/ConfigSoftwareVirtualize HP Procurve 5400zl ~$4000 for 48x1Gbps OF ver 1.0 + legacy 1 OF instance per VLAN NEC IP8800~Ask NEC for 48x1Gbps + 2x10 Gbps OF ver 1.0 + legacy 1 OF instance per virtual switch Toroki Lightswitch 4810 ~$4500 for 48x1Gbps + 2x10 Gbps OF ver 0.8.9 + legacy 1 OF instance per virtual switch Quanta LB4G with Stanford software ~$2500 for 48x1Gbps + 2x10 Gbps OF ver 1.0 (No legacy) 1 OF instance per switch

32. Other partners in the ecosystem OpenFlow Switch/router prototypes to come: – Dell, Netgear, Huawei Service Providers and carriers involved in deployments/trials: – Google, Internet2, JGN-Plus, NLR, Deutsche Telekom, British Telecom, France Telecom, NTT Further, there is active participation from: – BigSwitch, Broadcom, Brocade, Docomo, Ericsson, Extreme, Marvell …and many more!!

33. Ecosystem coming together Data Center Google, Amazon, Microsoft,.. NW Provider DT, DoCoMo (Level3, BT, Verizon,..) Data Center Google, Amazon, Microsoft,.. NW Provider DT, DoCoMo (Level3, BT, Verizon,..) Providers Researchers Research & Education Networks Researchers Research & Education Networks Academy Chip Vendors Switch/Router Vendors - Enterprise & Backbone - Packet & Circuit - Wireless New Class Chip Vendors Switch/Router Vendors - Enterprise & Backbone - Packet & Circuit - Wireless New Class Vendors (Hardware/Software) 33 Disclaimer: level of the interest differs

34. Exec Summary OpenFlow/SDN enables innovations within – Enterprise, backbone, & data center networks – Represents a promising architecture direction Providers like it for their own reasons – Enabling an ecosystem OpenFlow/SDN networks are being deployed – R&E networks around the world Time for regionals to get involved. Take advantage of GPO Solicitation 3.

35. Eight universities and two national research backbones OpenFlow as GENI Networking Substrate

36. OpenFlow Deployment at Stanford 36 OpenFlow-enabled buildings Gates Computer Science Building CIS EE Building Packard EE Building (soon) 20 switches from NEC, HP and Pronto 50 OpenFlow enabled WiFi APs WiMAX OpenFlow Service

37. KOREA OpenFlow Network Seoul Daejeon Deagu Busan Gwangju Suwon Controller VLAN on KOREN OpenFlow Switch (Linux PC) NOX OpenFlow Controller TJB TJB Broadcasting Company Japan OpenFlow Network Sapporo Studio Asahi Broadcasting Cooperation (ABC) at Osaka, Japan Sapporo Japan Server Data Transmission An Experiment of OpenFlow-enabled Network (Feb. 2009 - Sapporo Snow Festival Video Transmission) A video clip of Sapporo snow festival is transmitted to TJB (Daejeon, KOREA) via ABC server (Osaka, JAPAN).

38. Three New EU Projects: Ophelia, SPARC, CHANGE 38  L2 Packet  Wireless  Routing Pan-European experimental facility  L2 Packet  Optics  Content delivery  L2 Packet  Shadow networks  L2 L3Packet  Optics  Content delivery  L2 Packet  Emulation  Wireless  Content delivery

39. Current Trials 68 trials/deployments spanning 13 countries

40. Campuses and Backbones are participating… Time for regionals to deploy, experiment with and benefit from OpenFlow/SDN 40

41. Regional Opportunity Partner with Stanford, network researchers, and industry to bring innovations to your network and to your customers Deploy OpenFlow packet and circuit infrastructure Allow researchers to innovate on top Take control of your own infrastructure and innovate services as you need them 41

42. Regionals Well Positioned You have an infrastructure connecting key institutions strong relationships with vendors experience deploying experimental facilities a track record – deploying high end technologies in support of E-Science applications You have NOT enabled Internet architecture research as much and this is the time to do this 42

43. GPO Solicitation Calls for It 43

44. Interested? Contact Guru Parulkar (parulkar@stanford.edu) or Matt Davy (mpd@grnoc.iu.edu)parulkar@stanford.edumpd@grnoc.iu.edu 44