FINDing a GENI in a CCCastle

1
FINDing a GENI in a CCCastle

• Reinventing the Internet (and More) January
  2007
• Ellen W. Zegura

2
My role in GENI

• Member of GENI Research Coordination Working
  Group
• Member of CRA GENI Community Advisory Board
  (interim body)
• Member of CCC Interim Council (interim body)
• Thanks to Dave Clark, Peter Freeman, Guru
  Parulkar, many others for raw material

3
Outline

• A short story
• What is GENI?
• A research program (see FIND)
• A facility for experimenting at scale
• CCC and other acronyms
• Opportunities

4
Once upon a time

• There was a research project

ARPANET Logical Map, March 1977
5
That grew up to be the Internet.
6
Times changed, and it kept up
Everything on WEB
HTTP
IP
IP on everything
Ethernet
802.11
Satellite
Optical
Power lines
Bluetooth
ATM
Modified John Doyle Slide
7
And everyone (and thing) relied on it.
8
Then one day,

• People realized it wasnt working so well
• Lots of spam
• Phishing attacks to steal identity and more
• Too hard to set up my home network
• Cant tell why it isnt working
• Would you have tele-surgery over the Internet?
• (Why) is Google building their own network?
• And there were things it couldnt do
• Can you print my dream? (4 year old Bethany)

9
Scientists were heroic (and stymied)

• What is the Internet structure?
• How does it change and why?
• How robust is the Internet?
• What happens during a failure event?
• What are the properties of Internet paths?
• How do attacks propagate?
• What are users doing?

10
Even experts had to admit

• in the thirty-odd years since its invention,
  new uses and abuses, , are pushing the Internet
  into realms that its original design neither
  anticipated nor easily accommodates.
• Freezing forevermore the current architecture
  would be bad enough, but in fact the situation
  is deteriorating. These architectural
  barnaclesunsightly outcroppings that have
  affixed themselves to an unmoving architecture
  may serve a valuable short-term purpose, but
  significantly impair the long-term flexibility,
  reliability, security, and manageability of the
  Internet.
• Overcoming Barriers to Disruptive Innovation in
  Networking, NSF Workshop Report, 2005.

11
Along came the idea of a Future Internet
Future Internet (FI)E2E Networking and
Distributed Systems
Ethernet
802.11
Satellite
Optical
Power lines
Bluetooth
ATM
Link Technologies
Modified John Doyle Slide
12
Outline

• A short story
• What is GENI?
• A research program (see FIND)
• A facility/testbed for experimenting at scale
• CCC and other acronyms
• Opportunities

13
What is GENI?

• Global Environment for Networking Innovations
• Two parts
• The GENI Research Program, which will develop and
  evaluate ideas for the Future Internet
• The GENI Facility, which will provide the
  instrument for at-scale experimentation

14
GENI research program

• FIND U.S. National Science Foundation program to
  fund research on Future Internet Design (focus on
  architecture)
• Key areas of concern
• Security and robustness
• Mobility of hosts and networks
• Control and management
• Addressing, naming and (inter-domain) routing
• End-to-end principle vs in-network processing

15
Security trends

• Increasing vulnerabilities, viruses, attacks,
  worms
• 20 new vulnerabilities reported every day
• 120,000 known viruses and worms -- 50 new ones
  per day
• Large scale attacks doubling every year
• Increasing economic cost
• Viruses alone 60B
• Worldwide 105B
• Some ISPs have more than 90 traffic that is spam
• Identity thefts has emerged as a significant and
  serious threat
• And more

Source Spafford Talk
16
In the Future Internet

• Information disclosure control and integrity
• Important and well understood
• High availability -- suitable for even mission
  critical scenarios
• Balance of privacy and accountability
• Usable security for a range of users
• Context aware
• Different parts of the world have different
  requirements
• Appropriate for emerging devices and networks
• Mobile wireless and sensor networks, sensors,
  PDAs,
• Need coherent and comprehensive design

17
Mobile wireless trends

• 2B cell phones
• 400M cell phones with Internet capability --
  rising rapidly
• New data devices (blackberry, PDA, iPod) and
  services
• 240M vehicles on the road -- will soon get
  network connectivity with mobility
• Mobile computing and embedded devices to dominate
  future computing and communication
• Closed vertically integrated networks and services

18
In the Future Internet

• Seamless integration of networks
• Cellular WAN, wireless PAN, LAN and MANs, ad hoc
  mesh
• Build on current and new radio technologies
• 4G, WiMax, .11n, MIMO, cognitive radios, and more
  to come
• New protocol capabilities
• Cross-layer support, spectrum coordination,
  discovery, QoS, multi-hop
• Autonomic, self-
• Secured and privacy protecting
• Over otherwise shared wireless medium
• Accelerate new services and ability to deploy
  them
• location-aware, multimedia, dynamic communities

19
Snapshot of research challenges
20
Outline

• A short story
• What is GENI?
• A research program (see FIND)
• A facility for experimenting at scale
• CCC and other acronyms
• Opportunities

21
GENI facility motivation
Shared DeployedInfrastructure
Need for Large experimental facility/infrastructu
re
This chasm represents a majorbarrier to impact
real world
Small Scale Testbeds
Maturity
ResearchPrototypes
Foundations Research
Funded by current programs
Time
22
Link between research and facility

• Goal Seamless conception-to-deployment process

Deployment
Analysis
Simulation / Emulation
Experiment At Scale (Facility)
(models)
(code)
(results)
(measurements)
23
Facility goals

• Enable exploration of new network architectures,
  mechanisms, and distributed system capabilities
• A shared facility that allows
• Concurrent exploration of a broad range of
  experimental networks and distributed services
• Interconnection among experimental networks the
  commodity Internet
• Users and applications able to opt-in
• Observation, measurement, and recording of
  outcomes
• Help develop stronger scientific base

24
Facility design key concepts
Slicing, Virtualization, Programmability
25
Scope
26
Details of the Facility
Sensor Network
backbone wavelength
backbone switch
Internet
Customizable Router
Edge Site
Wireless Subnet
27
Outline

• A short story
• What is GENI?
• A research program (see FIND)
• A facility for experimenting at scale
• CCC and other acronyms
• Opportunities

28
What is the CCC?

• Computing Community Consortium
• Solicited by US NSF, calling for the computing
  research community to unite in the establishment
  of a Computing Community Consortium
• Serve as a community proxy responsible for
  facilitating the conceptualization and design of
  promising infrastructure-intensive projects
  identified by the computing research community to
  address compelling scientific grand challenges
  in computing.
• Initial responsibility would be guiding the
  design of the Global Environment for Network
  Innovations (GENI) on behalf of the research
  community, ensuring broad community participation
  in the GENI design process and identifying
  necessary pre-construction development
  activities.
• Award made to Computing Research Association
  (CRA)

29
And eventually there will be

• CCC Council
• Interim group appointed November 2006
• Nominations for members due late Jan 2007
• GENI Science Council (GSC)
• Interim group appointed October 2006
• Nominations for members in November 2006
• Initial permanent group to be named soon
• GENI Project Office (GPO)
• Solicitation in fall 2007
• Award due in spring 2007

30
Opportunities

• Researchers
• Contribute to research vision and agenda
• Engage in peer-to-peer collaborations and
  conversations about experiments
• ResearchersGovernment
• Industry

31
International Partnerships Important

• Help define facility scope
• Build national partner facilities to complement
  US GENI facilities and capabilities
• Share facilities with researchers in all partner
  countries
• Encourage collaborative international research
  projects and experiments

32
Industry Partnerships Important

• Help to refine RD objectives
• Become a member in the GENI consortium
• Provide leading-edge technology for use in GENI
• Contract (or subcontract) to build the facility
• Conduct collaborative research with universities
• Benefits to partnering
• Accelerate the transfer of academic research
  results to commercial products
• Enable a national/international proving ground
  for new technology

33
Conclusions

• The future of the Internet is too important to be
  left to chance or random developments.
• True experimentation is needed.
• The GENI project intends to provide the basic
  architectures, technologies, and policies that
  will be needed for successful networking in the
  2010-2020 time frame.

34
More Information

• Visit the GENI web site at
• http//www.nsf.gov/cise/geni/
• Visit the CISE Web site at
• http//www.nsf.gov/dir/index.jsp?orgCISE
• Visit the CRA CCC web site at
• http//www.cra.org/ccc

35
Acknowledgments

• The GENI Planning Group
• Peterson, Anderson, Blumenthal, Casey, Clark,
  Estrin, Evans, McKeown, Raychaudhuri, Reiter,
  Rexford, Shenker, Vahdat, Wroclawski
• The GENI Working Groups
• Research Coordination
• Facility Architecture
• Backbone
• Mobile wireless sensor networks
• Distributed services
• Planning grant workshops participants
• CISE GENI Team
• And others

36
Backup slides
37
Fitting parts together
38
Another Important Trend Networking the Physical
World
New Machines
39
Sensor Networking in Future Internet

• Sensor networks challenge Internet architecture
• host-to-host communication, addressing, routing,
  end-to-end principle,
• Sensor networks require
• Aggregate communication
• dissemination, data collection, aggregation
• Communication with data/logical services, not
  just devices
• Data centric as opposed to host centric
• Autonomic
• Self-configuration, self-management, self-
• Sensor networks constraints
• Limited resources, intermittent connectivity,
  mobility, in-network proc

40
Photonics Integration Trends
High capacity dynamic optical networks a
certainty
41
Future Internet and Dynamic Optical Networks
Circuit and PacketService Layer
All Optical Transport Core
How can Future Internet exploit an optical core
that can provide bandwidth on demand
dynamically with low latency and guarantees?
42
In-Network Processing Trends

• Middle boxes NAT, firewall, IDS, etc..
• WEB caching and content distribution networks
• Network based services computing, storage
• Internets end-to-end principle (a defining
  attribute) challenged and revisited