Trend: Networking Age

1
Trend Networking Age

• Sept. 2007

2
Trend Networking Age

• Virtual Schools
• Virtual Workplace
• Electronic commerce
• virtual enterprise
• new forms of value chains
• virtual Cash
• Internet entertainment
• interactive sitcom
• Ubiquitous

3
Trend Ubiquitous

4
Trend Convergence

• Telephone
• Voice Transport
• Cable TV
• Video Transport
• Computer
• Digital Media Storage/Handling
• News/Advertising
• Digital Media Production
• Merging of Content Providers and Content
  transporters
• Phone companies, cable companies, entertainment
  industry, and computer companies
• Convergence stages
• Networking / devices / consumer

5
Trend Information Glut

• Web gt
• Information production and dissemination costs
  are almost zero
• Too much information
• Needles in the haystack
• Thousands of hits on each search
• Need tools for summarizing the information
• Opportunities for artificial intelligence
• Need to express information so that both human
  and computers can understand

6
Networking Trends

• More Internet Traffic
• Data gt Voice (1998)
• Traffic gt Capacity ?
• Traffic Engineering
• Faster Media / Backbone
• Bandwidth
• Everything over IP
• NGI - NGN
• Wireless/Mobile 3G, 4G, Wi-Fi
• Ubiquitous

7
Trend More Internet Traffic

• Number of Internet hosts is growing
  super-exponentially
• Traffic per host is increasing
• LAN 1 Gbps
• Cable modems allow 1 to 10 Mbps access from home
• 6-27 Mbps over phone lines using ADSL/VDSL
• 11-54-100 Mbps Wireless LAN (Wi-Fi)
• Bandwidth requirements are doubling every 4 months

8
Trend Data gt Voice

9
Trend Traffic Engineering

• Users Performance Optimization
• gt Maximum throughput, Min delay, min loss, min
  delay variation
• Efficient resource allocation for the provider
• gt Efficient Utilization of all links
• gt Load Balancing on parallel paths
• gt Minimize buffer utilization
• Current routing protocols (e.g., RIP and OSPF)
  find the shortest path (may be over-utilized).
• QoS Guarantee Select paths that can meet QoS
• Enforce Service Level agreements
• Enforce policiesConstraint based routing - QoSR

10
Trend Faster Media / Backbone

• LAN 1 Gbps over 4-pair UTP-5 up to 100 m, 10G
  being discussed.
• Was 1 Mbps (1Base-5) in 1984.
• Wireless networks 54/100 Mbps (100m, 300m, 2km),
  
• 2.5 Gbps to 5km using light
• Was 1 Mbps (IEEE 802.11) in 1998.
• Backbone Dense Wavelength Division Multiplexing
  (DWDM)
• OC-768 40 Gbps, a to 65 km, 1.6 - 10
  Tbps.
• Was 100 Mbps (FDDI) in 1993.

11
DWDM - Dense Wavelength Division Multiplexing
2.488 Gbps (1)
1310/1510 nm
2.488 Gbps (16)
?1
?2
?3
?4
?5
?16
1530-1565 nm ramge
1310/1510 nm
162.488 Gbps 40 Gbps
16 uncorrelated wavelengths
16 stabilized, correlated wavelengts
12
Trend Access Bandwidth

13
Why Optical Networks? DWDM optoelectricl metro
network
14
Trend Wireless / Mobile

15
Trend Wi-Fi (Wireless Fidelity) in Your Future
(1)

16
Trend Wi-Fi in Your Future (2)

17
Trend Wi-Fi in Your Future (3)

18
Trend Wi-Fi in Your Future (4)

19
Integration of 3G and WLAN- offer possibility of
achieving anywhere, anytime, high speed and low
expense Internet access
3G WLAN
Wide area Local area
Low bit rate (2M when stand still) High bit rate (11M to 54M)
Data/Voice service (QoS support) Data service
High expense Low expense
High mobility Low mobility
20

21

22
Trend Everything over IP

• Data over IP gt IP needs Traffic engineering
• Voice over IP gt Quality of Service and Signaling
• Backbone -- Optic networksI
• IP and DWDM gt Winning combination
• IP for route calculation, traffic aggregation,
  protection
• DWDM gt Cheap bandwidth
• Avoid the cost of SONET/ATM equipmnt
• Internet technology ATM philosophy

23
Future Internet Research and Experimentation

• Oct. 2007

24
Todays Internet

• Millions of users
• Web, email, low-quality audio video
• Interconnect personal computers and servers
• Applications adapt to underlying technology
• Todays Internet Doesnt
• Provide reliable end-to-end performance
• Encourage cooperation on new capabilities
• Allow testing of new technologies
• Support development of revolutionary applications

25

26

27

28
Tomorrows Internet

• Billions of users and devices
• Convergence of todays applications with
  multimedia (telephony, video-conference, HDTV)
• Interconnect personal computers, servers, and
  embedded computers
• New technologies enable unanticipated
  applications (and create new challenges)

29
Vint Cerf Open Challenges

• My primary disappointment has been the slow pace
  of high speed access for residential customers
• The second area of disappointment is the slow
  uptake of version 6 of the Internet protocol
  (IPv6).
• Perhaps the third area is the continuing
  difficulty caused by viruses, worms and
  distributed denial of service attacks.

30
How to make the Internet better???!

• Addressing current problems
• Security??
• Privacy??
• Self-diagnosis self-healing networks??
• Cheap connectivity for poor area and third world
  countries??
• Wireless mesh networks??
• sensors??
• Mobility??
• New cool apps??
• What is after IPTV, VoIP, BitTorrent,

31
Why Internet2 (1996)?

• The Internet was not designed for
• Millions of users
• Congestion
• Multimedia
• Real time interaction
• But, only the Internet can
• Accommodate explosive growth
• Enable convergence of information work, mass
  media, and human collaboration

32
Internet2 Project

• Develop and deploy advanced network applications
  and technologies, accelerating the creation of
  tomorrows Internet.
• Goals
• Enable new generation of applications
• Re-create leading edge RE network capability
• Transfer capability to the global production
  Internet
• 206 University Members, Jan. 2005

33
Internet2 Focus Areas

• Advanced Network Infrastructure - Abilene
• Backbones operate at 10 Gbps capacity - 100 Gbps
  (2007)
• Middleware
• A layer of software between the network and the
  applications
• Authentication
• Identification
• Authorization
• Directories
• Security
• Engineering
• IPv6, Measurement, Multicast, QoS, Routing,
  Security, Topology
• Advanced Applications
• Distributed computation
• Virtual laboratories
• Digital libraries
• Distributed learning
• Digital video
• Tele-immersion
• All of the above in combination

34
Abilene Connections July 2006
Abilene Connections Apr-2000
35
Abilene International Peering 2006
Abilene Connections Apr-2000
36
Internet Development Spiral
Commercialization
Privatization
Todays Internet
Internet2
Research and
Development
Partnerships
Source Ivan Moura Campos
37
Internet2 and the Next Generation Internet
Initiative
Internet2
NGI
Federal agency-led
University-led
Developing education and research driven
applications
Agency mission-driven and general purpose
applications
Building out campus networks, gigaPoPs and
inter-gigapop infrastructure
Funding research testbeds and agency research
networks
Interconnecting and interoperating to provide
advanced networking capabilities needed to
support advanced research and education
applications
38

39

40

41

42
The Current Internet Sucks!

• The Internets current evolutionary path will
  not adequately meet our future needs
• Security
• Reliability
• New application and user requirements
• ..
• We will need a significant architectural change
• Perhaps not now, but eventually

43
The Current Internet Research Sucks!

• The mid 80s up to mid 90s
• Researchers invented the Internet
• Had tremendous influence in the early days of
  the commercial Internet
• mid 90s to present
• IPv6, multicast, QoS researchs last
  successes/failures
• Since then totally irrelevant architecturally
• Helped build better routers, etc.
• But no impact on basic architecture
• Simultaneous with shift from industrial labs to
  academia???

44
Our Founding (Funding) Fable

• Researchers invent new architectures
• Architectures are validated on a testbed
• IETF, ISPs, and router vendors collaborate to
  deploy new design
• This is complete BS!

45
Do Traditional Testbeds Really Test?

• Production-oriented testbeds
• Real traffic provides good validation
• But can test only very incremental changes
• Research-oriented testbeds
• Can test radical architectures
• Lack of real traffic results in poor validation
• Both are expensive (dedicated bandwidth)

46
What about Deployment?

• Architectural change requires ISP consensus
• - Hard to agree
• - No competitive advantage from architectural
  innovation
• - All have huge sunk investment in the status quo
• ISPs are unlikely candidates for architectural
  change
• Architecture isnt just static, its decaying
• Ad hoc workarounds muddy the architectural waters

47
We are at an Impasse

• We cant test new architectures
• - Despite sizable investments in testbeds
• We cant deploy new architectures
• - And things are getting worse, not better
• Yet there are pressing requirements for which
  the current architecture is not well suited

48
The Communitys Response

• Focus on areas where we can have impact
• Empirical studies
• Incremental changes (subject to current
  constraints)
• Small stream of architectural proposals
• Paper designs without hope of deployment
• More science fiction than engineering
• Have largely abandoned hope of effecting
  fundamental architectural change
• Living with, rather than overcoming, the impasse

49
Overcoming the Impasse?

• Must be able to test new architectures
• Wide range of architectures
• Real traffic from willing individuals
• Low overhead for individual researchers
• Meet the grand challenge of reinventing the
  Internet
• Empirical, incremental research is great, but
  not enough
• If someone put us in charge, what would we do?
• What about deployment?
• Several options, none good, but no excuse to not
  have an answer to the grand challenge

50
Testing Virtual Testbed

• Overlay testbed (think RON, etc.)
• Host proxy directs packets to overlay
• Proxy must architecturally neutral, and flexible
• Individuals (anywhere) opt-in by turning on
  proxy
• Shared testing infrastructure (think Planetlab)
• Overlay nodes shared among experiments
• Slicing on per-packet timescales
• Virtualized routers
• These ideas have turned into the GENI program

51
Why GENI (Global Environment for Network
Innovations) / Future Internet Network (2005)?

• The original designing idea of current
  information networks is basically a specific
  network supports one major service.
• The limitation of the original designing idea can
  not support the multiple requirements for
  networks and services (such as Telecom
  network?Internet)
• The original design mode of Internet leads to
  its shortage in mobility, security, controllable
  and administrable.
• We urgently need redesign the framework of new
  generation Internet to overcome the serious
  shortage of current Internet.
• Providing pervasive and trusted services based on
  a specific network has became the key research
  directions of national informatization.

52
GENI (Global Environment for Network
Innovations) - NSF 2005 (1)

• What is GENI?
• GENI is a facility concept being explored by the
  US computing community
• back to an NSF workshop in 2005
• focus on architectural research, and provide the
  experimental infrastructure needed to support
  that research
• focus on the research agenda (and infrastructure
  needs) of the optical, wireless, sensor network,
  and distributed systems communities

53
GENI (Global Environment for Network
Innovations) (2)

• The goal of GENI
• Goal a Future Internet that meets the demands
  of 21st century
• to increase the quality and quantity of
  experimental research outcomes in networking and
  distributed systems
• to accelerate the transition of these outcomes
  into products and services
• enhance economic competitiveness and secure the
  Nation's future
• Ultimately, to lead to a transition of the
  Internet

54
GENI Research Opportunities (3)

• Two dichotomy thought (false, or at least
  unnecessary)
• a "clean slate" reconceptualization of Internet
  architecture
• today's 30-year-old architecture that limit its
  ability to cope with emerging threats and
  opportunities
• eroding trust, reduced innovation, slowing
  update
• future innovation will take the form of new
  services and applications running on top of the
  Internet

55
GENI Research Opportunities (4)

• GENI discussion
• First, we interpret "Future Internet" very
  broadly to include innovations at any level of
  the architecture
• alternative protocols and architectures running
  inside the network (as overlays on top of today's
  network)
• Second, research should employ clean slate
  thinking, but this does not imply that an
  entirely new Internet will be necessary. In other
  words, "clean slate" is a process, not a result.
• Third, opportunities between two perspectives
• exploring how today's architecture is best
  evolved to support emerging overlay services.

56
GENI Facility Concept (5)

• Facility Concept
• experimental platforms for both research and
  deployment
• filling the gap between small-scale exp and
  mature tech
• GENI evaluate new network systems on large-scale
• Two levels
• Physical level, GENI substrate will consist of a
  collection of links, forwarders, storage,
  processors, and wireless net
• On top of this substrate, a software management
  framework will be overlay network experiments on
  the substrate

57
GENI Facility Concept (6)

• Four key ideas
• substrate components will be programmable -
  possible to embed any network experiment,
  including clean-slate designs
• Substrate will be virtualizable - possible to
  embed multiple slices in the substrate at the
  same time (allow experimental services and
  architectures to run continuously)
• GENI will include mechanisms that allow
  end-users to seamlessly opt-in to experimental
  services
• GENI will be modular (architecture and
  interfaces) - possible to extend GENI with new
  networking technologies . GENI will not be a
  static artifact, but rather a dynamic
  infrastructure that is continually renewed.

58
FIND (Future Internet Network Design) NSF 2006
(1)

• FIND asks two broad questions
• What are the requirements for the global network
  in 15 years
• How would we re-conceive tomorrow's global
  network today, if we could design it from
  scratch?
• FIND program solicits "clean slate process"
  research proposals in the broad area of network
  architecture, principles, and design

59
FIND (Future Internet Network Design) (2)

• FIND research might address wide questions
• What will the edge of the network look like in
  15 years? How might the network architecture of
  15 years hence best accommodate sensors, embedded
  systems, and the like?
• How might the network of 15 years from now
  support what users really do (and care about)?
  How might such functions as information access,
  location management or identity management best
  fit into a new overall network architecture?
• What will the core of the network look like in
  15 years? How might the changing economics of
  optical systems affect the overall design of the
  larger network?

60
Clean Slate Network (1)

• 100x100 Clean Slate Project - NSF November 2003
• CMU, Fraser Research, Stanford, Berkeley, Rice,
  ATT Research, Internet 2
• Clean Slate Network Stanford 2005
• They believe that the current Internet
• has significant deficiencies that need to be
  solved
• Internet's shortcomings will not be resolved by
  the conventional incremental and
  'backward-compatible' style
• Program can be characterized by two research
  questions
• Today, if we were to start again with a clean
  slate, how would we design a global
  communications infrastructure?
• How should the Internet look in 15 years?

61
Clean Slate Network (2)

• Five key areas for research
• Network architecture
• Heterogeneous applications
• Heterogeneous physical layer technologies
• Security
• Economics policy
• Research projects
• Flow Level Models for the Future Internet
• Clean Slate Approach to Wireless Spectrum Usage
• Fast Dynamic Optical Light Paths for the
  Internet core
• A Clean Slate Approach to Enterprise Network
  Security

62
FIRE (Future Internet Research and
Experimentation) European 2007 (1)

• Internet has grown to an unexpected reach, as
  for the number of users, capacity of the links,
  broadband penetration to the home, services.
• Problems spam, viruses, denial of service
  attacks, complexity of management
• Internet drawbacks and limitations to
  scalability, suitability, mobility, transparency,
  security
• require new radical approaches to fundamentally
  redesign its protocols and architectures.

63
FIRE (2)

• FIRE is an experimentally-driven long-term
  research initiative on Future Internet concepts,
  protocols and architectures
• encompassing technological, industrial and
  socio-economic aspects
• acting as proof-of-concept of the newly proposed
  technologies and services
• FIRE RESEARCH
• Long term multidisciplinary research on future
  internet paradigms
• Open to fresh bottom-up ideas with no
  backwards-compatibility constraints
• Building on the FET SAC initiative Situated and
  Autonomic Communications
• Considering at the same time technological,
  economic and social/policy aspects
• Build in from the outset and on all levels the
  right balance between security / accountability
  and privacy

64
FIRE (3)

• FIRE EXPERIMENTATION
• Large scale experimentation of new paradigms and
  concepts for the future internet and related
  service architectures
• Learning through broad experimentation
• Integrating and validating new concepts
• Federating and extending existing testbeds and
  research infrastructures
• no backwards-compatibility constraints
• European approach and develop a European identity

65
Looking Over the Fence at Networking

• Jennifer Rexford

66
Internet Success Leads to Ossification

• Intellectual ossification
• Pressure for backwards compatibility with
  Internet
• Risks stifling innovative intellectual thinking
• Infrastructure ossification
• Limits on the ability to influence deployment
• E.g., multicast, IPv6, QoS, and secure routing
• System ossification
• Shoe-horn solutions that increase system
  fragility
• E.g., NATs (network address translation) and
  firewalls

67
A Need to Invigorate Networking Research

• Measurement
• Understanding the Internet artifact
• Better built-in measurement for the future
• Modeling
• Performance models faithful to Internet realities
• X-ities like manageability, evolvability,
  security,
• Prototyping
• Importance of creating disruptive technology
• Emphasis on enabling new applications

68
Modeling The X-ities (or Ilities)

• Beyond higher speed to consider X-ities
• Reliability
• Scalability
• Manageability
• Configurability
• Predictability
• Non-fragility
• Security
• Evolvability
• Challenging to model, or even to quantify

69
A Need for Interdisciplinary Work

• Statistical analysis
• Artificial intelligence
• Maximum likelihood estimation
• Streaming algorithms
• Cryptography
• Optimization
• Information theory
• Game theory and mechanism design

70
Discussion

• Where should the intelligence reside?
• Traditional Internet says the edge
• What about middle boxes (e.g., NAT)?
• Need to assemble applications from components
  located in different parts of the network?
• Better isolation and diagnosis of faults?
• Decentralized Internet makes this difficult
• Need to detection, diagnosis, and accountability
• Challenges the end-to-end argument

71
Discussion

• Data as a first-class object?
• Tradition Internet simple moves the bytes
• Naming, search, location, management in the net
• Modifying the data as it traverse the network
• Does the Internet have a control plane?
• Traditional Internet stress data transport
• What about network management and control?
• Today we place more emphasis on designing new
  protocols and mechanisms than controlling them

72
Discussion

• Abstractions on topology and performance
• Traditional Internet hides details from end hosts
• Network properties are, at best, inferred
• Guidelines for placement of middle boxes?
• Feedback info about topology and performance?
• Beyond cooperative congestion control
• Traditional Internet places congestion control in
  the end hosts, and trusts them to behave
• Is this trust misguided?
• New alternatives to congestion control?

73
Discussion

• Incorporating economic factors in design
• Traditional Internet ignores competitive forces
• Many constraints are economic, not technical
• Better to construct/align economic incentives
• Ways to deploy disruptive technology
• Traditional core is not open to disruptive tech
• Overlay network as a deployment strategy
• Other approaches? Virtualization? Middle boxes?
  Speaking the legacy protocols with new logic?
• Experimental facilities? A do over?

74
The Innovators Dilemma

• Leading companies often miss next big thing
• E.g., disk-drive industry and excavation
  equipment
• Problem
• Listening to customers leads to incremental
  improvement on the existing technology curve
• Disruptive technologies are often less effective
  for the existing customers, so tend to be ignored
• New companies exploit the new technology for a
  new market (e.g., desktops, laptops)
• Eventually, the new technology curve overtakes
  the old technology, usurping the old technology
• Will this happen with the Internet?