1. Software Defined Networking: tecnologia e prospettive
Prof. Stefano Salsano
Seminario nel corso “Advanced Networking and Internet Modeling„ Prof. Francesco Lo Presti
26 Maggio 2015

2. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation SDN approach, goals and dreams… A bit of technology: OpenFlow Application examples SDN and cloud Google’s SDN WAN SDN and Network Function Virtualization

3. Internet success
The Internet success is a remarkable story, from a research infrastructure to a global network, interconnecting billions of devices and people
Innovation looks easy on the Internet as we witness always new and more powerful services and applications
Web, P2P, VoIP, social networks, video streaming…

4. Network ossification The history is a bit different behind the scene:
Huge complexity
Few people can innovate
Closed equipment
Network «ossification» 4 4

5. Classical network architecture
Distributed control plane
Distributed routing protocols: OSPF, IS-IS, BGP, etc.
Feature
Operating
System
Feature
Specialized Packet Forwarding Hardware
Operating
System
Feature
Specialized Packet Forwarding Hardware
Operating
System
Classical computer network architecture
Feature
Specialized Packet Forwarding Hardware
Operating
System
Specialized Packet Forwarding Hardware
Feature
Operating
System
Specialized Packet Forwarding Hardware

6. The Networking Industry (2010s)
Routing, management, mobility management, access control, VPNs, …
Feature
Feature
Million of lines of source code
5400 RFCs
Barrier to entry
Operating
System
Specialized Packet Forwarding Hardware
Billions of gates
Complex
Power Hungry
The networking industry is broken. Why doesn’t it look like the server market? Historical accident.
Overwhelming complexity.
The history of OpenFlow starts with a realization that the state of the industry is broken.
Analogy: like the computer industry in the late 70’s and early 80’s before IBM compatibles and Windows: vertically integrated.
Closed, vertically integrated, boated, complex, proprietary
Many complex functions baked into the infrastructure
OSPF, BGP, multicast, differentiated services, Traffic Engineering, NAT, firewalls, MPLS, redundant layers, …
Little ability for non-telco network operators to get what they want
Functionality defined by standards, put in hardware, deployed on nodes 6 6

7. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation SDN approach, goals and dreams… A bit of technology: OpenFlow Application examples SDN and cloud Google’s SDN WAN SDN and Network Function Virtualization

8. Software Defined Network
Well-defined open API
Constructs a logical map of the network
Feature
Feature
Northbound
Network OS
Open, vendor-agnostic protocol
Southbound
OpenFlow
Simple Packet
Forwarding
Hardware
Simple Packet
Forwarding
Hardware
Simple Packet
Forwarding
Hardware
Simple Packet
Forwarding
Hardware
Simple Packet
Forwarding
Hardware
17/09/2013

9. Analogy with IT industry: from mainframes to PCs
App
Specialized
Applications
Linux
Mac OS
Windows
(OS) or
Specialized
Operating
System
Specialized
Hardware
Microprocessor
The mainframe industry in the 1980s was vertical and closed: it consisted of specialized hardware, operating system and applications --- all from the same company.
A revolution happened when open interfaces started to appear. The industry became horizontal. Innovation exploded.
Mainframe industry in the 1980s: Vertically integrated
Closed, proprietary
Slow innovation
Small industry
Horizontal
Open interfaces
Rapid innovation
Huge industry

10. Analogy with IT industry: from closed box to SDN
Specialized
Features
App
Specialized
Control
Plane
Control
Plane
Control
Plane
Control
Plane or or
Specialized
Hardware
Merchant
Switching Chips
Networking industry in 2010s: Vertically integrated
Closed, proprietary
Slow innovation
Horizontal
Open interfaces
Rapid innovation

11. SDN Concept Separate Control plane and Data plane entities
Network intelligence and state are logically centralized
The underlying network infrastructure is abstracted from the applications
Execute or run Control plane software on general purpose hardware
Decouple from specific networking hardware
Use commodity servers
Have programmable data planes
Maintain, control and program data plane state from a central entity
An architecture to control not just a networking device but an entire network

12. Network OS and OpenFlow
Distributed system that creates a consistent, up-to-date network view, runs on servers (controllers) in the network
Uses an open protocol to:
Get state information from forwarding elements
Give control directives to forwarding elements
OpenFlow
is a protocol for remotely controlling the forwarding table of a switch or router
is one element of SDN

13. Abstractions in the Control Plane
Network Virtualization
Well-defined API
Routing
Traffic Engineering
Other Applications
Network Map
Abstraction
Network Operating System or “Controller”
Forwarding
Separation of Data
and Control Plane
Forwarding
Forwarding
Forwarding

14. Forwarding Abstractions
Purpose: Abstract away forwarding hardware
Flexible
Behavior specified by control plane
Built from basic set of forwarding primitives
Minimal
Streamlined for speed and low-power
Control program not vendor-specific
OpenFlow is an example of such an abstraction

15. SDN promises (or dreams…)
Innovation
beyond IP, clean slate approaches…
Change of paradigm «Redefining Abstractions» (see Scott Shenker presentation)
Openness
open fast switching hardware, open controllers

16. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation SDN approach, goals and dreams… A bit of technology: OpenFlow Application examples SDN and cloud Google’s SDN WAN SDN and Network Function Virtualization

17. Traditional network node: Switch
Typical Networking Software
Management plane
Control Plane – The brain/decision maker
Data Plane – Packet forwarder

18. Traditional network node: Router
Router can be partitioned into control and data plane
Management plane/ configuration
Control plane / Decision: OSPF (Open Shortest Path First)
Data plane / Forwarding
Adjacent Router
Router
Management/Policy plane
Configuration / CLI / GUI
Static routes
Control plane
OSPF
Neighbor table
Link state database
IP routing table
Forwarding table
Data plane
Routing
Switching

19. OpenFlow Controller Control Path OpenFlow Data Path (Hardware)
OpenFlow Basics
OpenFlow Controller
OpenFlow Protocol (SSL/TCP)
Control Path
OpenFlow
Data Path (Hardware)

20. OpenFlow Basics Network OS Ethernet Switch Control Program A
Control Program B
Network OS
OpenFlow Protocol
Ethernet Switch
Control Path
OpenFlow
Data Path (Hardware)

21. OpenFlow Basics Network OS Control Program A Control Program B
“If header = p, send to port 4”
“If header = q, overwrite header with r, add header s, and send to ports 5,6”
Packet
Forwarding
“If header = ?, send to me”
Flow
Table(s)
Packet
Forwarding
Packet
Forwarding

22. OpenFlow Primitives <Match, Action>
Match arbitrary bits in headers:
Match on any header, or new header
Allows any flow granularity
Action
Forward to port(s), drop, send to controller
Overwrite header with mask, push or pop
Forward at specific bit-rate
<Match, Action>
Header
Data
Match: 1000x01xx x

23. General Forwarding Abstraction
Small set of primitives
“Forwarding instruction set”
Protocol independent
Backward compatible
Switches, routers, WiFi APs, basestations, TDM/WDM

24. OpenFlow example OpenFlow Client Controller PC Software Layer MAC src
Flow Table
MAC
src
dst IP
Src
Dst
TCP
sport
dport
Action
Hardware
Layer *
port 1
port 1
port 2
port 3
port 4

25. OpenFlow Basics Flow Table Entries
Rule
Action
Stats
Packet + byte counters
Forward packet to zero or more ports
Encapsulate and forward to controller
Send to normal processing pipeline
Modify Fields
Any extensions you add!
Now I’ll describe the API that tries to meet these goals.
Switch
Port
VLAN ID
VLAN
pcp
MAC
src
MAC
dst
Eth
type IP
Src IP
Dst IP
ToS IP
Prot L4
sport L4
dport
+ mask what fields to match

26. Examples Switching Flow Switching Firewall Switch Port MAC src dst Eth
type
VLAN ID IP
Src
Dst
Prot
TCP
sport
dport
Action * *
00:1f:.. * * * * * * *
port6
Flow Switching
Switch
Port
MAC
src
dst
Eth
type
VLAN ID IP
Src
Dst
Prot
TCP
sport
dport
Action
port3
00:20..
00:1f..
0800
vlan1 4
17264 80
port6
Firewall
Switch
Port
MAC
src
dst
Eth
type
VLAN ID IP
Src
Dst
Prot
TCP
sport
dport
Action * * * * * * * * * 22
drop

27. Reactive vs. Proactive (pre-populated)
First packet of flow triggers controller to insert flow entries
Efficient use of flow table
Every flow incurs small additional flow setup time
If control connection lost, switch has limited utility
Extremely simple fault recovery
Proactive
Controller pre-populates flow table in switch
Zero additional flow setup time
Loss of control connection does not disrupt traffic
Essentially requires aggregated (wildcard) rules
Simple fault recovery but not necessarily simple or optimal

28. Microflow vs. Aggregated
Every flow is individually set up by controller
Exact-match flow entries
Flow table contains one entry per flow
Good for fine grain control, policy, and monitoring, e.g. campus
Aggregated
One flow entry covers large groups of flows
Wildcard flow entries
Flow table contains one entry per category of flows
Good for large number of flows, e.g. backbone

29. Virtualization or “Slicing”
Virtualization trend
Controller 1
App
Controller 2
Virtualization or “Slicing”
OpenFlow
NOX
(Network OS)
Network OS
App
App
App
Windows
(OS)
Linux
Mac OS
Windows
(OS)
Linux
Mac OS
Windows
(OS)
Linux
Mac OS
Virtualization layer
x86
(Computer)
Computer Industry
Network Industry

30. Virtualization or “Slicing” Layer
Isolated “slices”
Many operating systems, or
Many versions
App
Network Operating System 1
Network Operating System 2
Network Operating System 3
Network Operating System 4
Open interface to hardware
Virtualization or “Slicing” Layer
Open interface to hardware
Simple Packet Forwarding Hardware
Simple Packet Forwarding Hardware
Simple Packet Forwarding Hardware
Simple Packet Forwarding Hardware
Simple Packet Forwarding Hardware
17/09/2013 30

31. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation SDN approach, goals and dreams… A bit of technology: OpenFlow Application examples SDN and cloud Google’s SDN WAN SDN and Network Function Virtualization

32. SDN and cloud
Cloud
x 10000… “multi-tenant” multi site

33. SDN and cloud
Cloud computing service providers face the issue of multi-tenancy at the network level
IP and Ethernet each have virtual network capability, but limited in terms of
how many tenants can be supported
how isolated each tenant
configuration and management complexity
SDN is increasingly accepted as the path to "cloud networking"

34. Neutron service in the OpenStack cloud architecture (was Quantum)
Nova Compute
Service
Virtual Machines
Swift
Storage
Object Store
Basic Network Connectivity
Nova, Swift, and Neutrum API
Servers
Disks
Neutrum
Service
Virtual Networks
Networks
Developers have ability to create multiple networks for their own purposes (multi-tier apps)
May support provisioning of both virtual and physical networks – differences captured through plugin’s

35. Neutrum service in the OpenStack cloud architecture
User Application – CLI - Horizon Dashboard - Tools
Tenant API
Tenant API
Network Service (Neutrum)
Compute Service
(Nova)
System
Admin
Internal API
Admin API
Plug-In
Compute Node
Hypervisor vSwitch
Physical
Network Router/Switch
Clustered Network Controller

36. Available Neutrum plug-ins
Open vSwitch
Linux bridge
Nicira NVP
Cisco (Nexus switches and UCS VM-FEX)
WIP: VXLAN
NTT Labs Ryu OpenFlow controller
NEC OpenFlow
Big Switch Floodlight
Most of them are SDN based !

37. The zoo of network technologies

38. Data center simplification
Traditional Network
Fabric

39. Network Fabric Defined
Flat network
Every port virtually connected to every other
High speed network; 10 Gig-E, roadmap to 40 Gig-E and 100 Gig-E
Operationally simple
Optimized for virtual traffic and east west traffic flows
Optimized packet processing

40. The shift toward SDNs
Allows for the network to be in better alignment with the current data center trends
Brings a high level of agility to the network
Enables programmability
Improves application performance
Abstracts the control layer from the network infrastructure lay
Complements fabrics
Creates scalable network virtualization

41. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation SDN approach, goals and dreams… A bit of technology: OpenFlow Application examples SDN and cloud Google’s SDN WAN SDN and Network Function Virtualization

42. Google’s B4: an SDN success story
In April 2012 Google presented their OpenFlow based WAN (“B4”), globally interconnecting the Data Centers
B4 is based on a SDN architecture using OpenFlow to control relatively simple switches built from merchant silicon
Google engineered the switches and the SDN architecture

43. B4 worldwide deployment (2011)

44. Typical WAN engineering rules
WAN links are typically provisioned to 30-40% average utilization. This can mask virtually all link or router failures from clients.
Such overprovisioning give reliability at the costs of 2-3x bandwidth over-provisioning and high-end routing gear.

45. Google’s Requirements
Google fully controls applications and server using the WAN
The most bandwidth intensive applications are large-scale data copies: they can adapt to available capacity
The number of data center is limited, making centralized bandwidth control feasible

46. Traffic Engineering
Using SDN, B4 simultaneously supports standard routing protocols and centralized TE
TE algorithms allow to:
adjudicate among competing demands during resource constraint
use multipath forwarding/tunneling
dynamically reallocate bandwidth in the face of link/switch failures
Many B4 links to run at near 100% utilization and all links to average 70% utilization

47. The SDN switches

48. SDN based WAN architecture

49. Performance measurements
link utilization
ratio high prio / low prio
low prio loss
high prio loss

50. …and if you are not Google ?
A truly open SDN eco-system will grow, including controllers/network OS, management tools, switching equipment, and “network applications” (e.g. a TE component) Vendors will include SDN concepts in their solutions, mostly in a proprietary way ?

51. «SDN washing»
All main networking equipment vendors are now offering their SDN products and solutions
“SDN washing” – when networking vendors essentially take their existing technologies and try to re-label them as SDN products

52. OpenDayLight
Cisco, Juniper, Ericsson, IBM, NEC and other network vendors are joining up to standardize SDN (April 2013) with OpenDayLight project

53. OpenDayLight

54. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation SDN approach, goals and dreams… A bit of technology: OpenFlow Application examples SDN and cloud Google’s SDN WAN SDN and Network Function Virtualization

55. SDN and NFV (Network Function Virtualization)
NVF: a Network-operator-driven specification group within ETSI.
Initiated by 13 carriers now grown to 23 members

56. SDN and NFV (Network Function Virtualization)
Independent
Software Vendors
BRAS
Firewall
DPI
CDN
Tester/QoE
monitor
WAN
Acceleration
Message
Router
Radio Network
Controller
Carrier Grade NAT
Session Border
Classical Network Appliance Approach
PE Router
SGSN/GGSN
Generic High Volume
Ethernet Switches
Generic High Volume Servers
Generic High Volume Storage
Orchestrated,
automatic
remote install
Network Functions Virtualisation Approach
hypervisors
The Classical network appliance approach uses a proprietary chassis for every new service, some typical examples are shown above left. Although inside the appliances use similar chips they are packaged very differently outside. This creates complexity through the entire life cycle of network services e.g. Design, procurement, test, deployment, configuration, repair, maintenance, replacement, end-of-life, removal. There is no economies of scale (some boxes may only sell in the thousands as opposed to 9 Million IT servers every year). The need for start-ups to develop new hardware, including getting it NEBs and ETSI qualified, is a significant cost and deterrent to enter the telecoms market.
The Network Virtualisation (NV) approach replaces physical network appliances with software virtual appliances running on commodity IT servers. Using open IT techniques allows a competitive and innovative ecosystem to exploit the many x86 and Linux programmers in the world. The Virtual Appliances can be installed on IT servers using orchestration software, this will automatically and remotely install software, driven either by traffic demands or customer orders. To complement the standard high volume IT servers we will also use standard high volume storage and Ethernet switches. The standard high volume Ethernet switches will use merchant silicon which spreads the cost of switching ASICs over the widest possible Enterprise & Carrier market. 56

57. NFV and SDN are complementary
Open
Innovation
Network
Functions Virtualisation
Software
Defined
Networks
Creates operational flexibility
Reduces Reduces CapEx, OpEx, space & power delivery time consumption
Creates control abstractions to foster innovation.
Creates competitive supply of innovative applications by third parties

58. Outline
SDN motivations: Internet ossification, network complexity, barriers to innovation
SDN approach, goals and dreams…
A bit of technology: OpenFlow
Application examples
SDN and cloud
Google’s SDN WAN
SDN and Network Function Virtualization
Final remarks
17/09/2013

59. The two paths to SDN (r)evolution
New Abstractions / high level modeling of networks
Disruptive low cost net architectures and open hardware
Open Source Software (for Carriers’ Class nodes)
Open Innovation
Revolutionary path : disruptive innovation
Evolutionary path : progressive innovation
Seamless integration in current networks, compatibility with legacy
Solutions from traditional Vendors (or even Start-ups) …
Costs Reductions (CAPEX, OPEX)

60. Questions ? Thank you for your attention ! Stefano Salsano, Ph. D.
Associate Professor
Phone:
Fax:
UNIVERSITY OF ROME “TOR VERGATA”
Department of Electronics Engineering
Via del Politecnico, Rome - Italy

61. Suggested Readings
A. Manzalini, V. Vercellone, M. Ullio, «Software Defined Networking: sfide ed opportunità per le reti del futuro», Notiziario Tecnico Telecom Italia, n.1/2003
N. McKeown et al. «OpenFlow: Enabling Innovation in Campus Networks», CCR

62. Main sources Jennifer Rexford “Enabling Innovation Inside the Network”
Scott Shenker with Martín Casado, Teemu Koponen, Nick McKeown and others “The Future of Networking, and the Past of Protocols”
Rob Sherwood (with help from many others) “An Experimenter’s Guide to OpenFlow” - GENI Engineering Workshop June 2010
Brandon Heller, Rob Sherwood, David Erickson, Hideyuki Shimonishi, Srini Seetharaman, Murphy McCauley, “Tutorial 1: SDN for Engineers”
Dan Pitt, “The Open Networking Foundation: OpenFlow & SDN from lab to market”
Yeh-Ching Chung, “Network Virtualization - Software Defined Network”

63. Main sources
Tom Nolle “The role of software-defined networks in cloud computing”
Lew Tucker, “Quantum: What it is and Where it’s going”
Zeus Kerravala, “The Time for ICT is Now”
Tom Nollle “Understanding the relationship between SDN and NFV”
Bob Briscoe (+ Don Clarke, Pete Willis, Andy Reid, Paul Veitch), “Network Functions Virtualisation”
Antonio Manzalini, “Software Will Eat the Networks – Welcome to the Blue SDN”

64. My work on SDN
OSHI Open Source Hybrid IP/SDN networking
The DREAMER project
S. Salsano, P. L. Ventre, F. Lombardo, G. Siracusano, M. Gerola, E. Salvadori, M. Santuari, M. Campanella, L. Prete “OSHI - Open Source Hybrid IP/SDN networking and Mantoo - a set of management tools for controlling SDN/NFV experiments”, submitted paper (May 2015)
S. Salsano, P. L. Ventre, L. Prete, G. Siracusano, M. Gerola, E. Salvadori, “Open Source Hybrid IP/SDN networking (and its emulation on Mininet and on distributed SDN testbeds)”, 3rd European Workshop on Software Defined Networks, EWSDN 2014, 1-3 September 2014, Budapest, Hungary
M. Gerola, M. Santuari, E. Salvadori, S. Salsano, P. L. Ventre, M. Campanella, F. Lombardo, G. Siracusano, “ICONA: Inter Cluster ONOS Network Application”, demo paper, 1st IEEE Conference on Network Softwarization (Netsoft 2015), London, UK, April 2015
N. Blefari-Melazzi, A. Detti, G. Morabito, S. Salsano, L. Veltri, “Information Centric Networking over SDN and OpenFlow: Architectural Aspects and Experiments on the OFELIA Testbed”, to appear in Elsevier Computer Networks, Special Issue on Information-Centric Networking (ICN), 2013
N. Blefari-Melazzi, A. Detti, G. Mazza, G. Morabito, S. Salsano, L. Veltri, “An OpenFlow-based Testbed for Information Centric Networking”, Future Network & Mobile Summit 2012, 4-6 July 2012, Berlin, Germany
L. Veltri, G. Morabito, S. Salsano, N. Blefari-Melazzi, A. Detti, “Supporting Information-Centric Functionality in Software Defined Networks”, SDN’12: Workshop on Software Defined Networks, Co-located with the IEEE International Conference on Communications (ICC), June , Ottawa, Canada
A. Detti , C. Pisa, S. Salsano, N. Blefari-Melazzi, “Wireless Mesh Software Defined Networks (wmSDN)”, The 2nd International Workshop on Community Networks and Bottom-up-Broadband (CNBuB 2013), Lyon, France, October 7th, 2013
17/09/2013