Re-inventing the Internet

Vint Cerf

September 2012
1

The Original ARPANET Dec 1969

940

#2 SRI #3 UCSB #1 UCLA Sigma 7

#4 UTAH

360

PDP 10

Internet 1999

Internet - Global Statistics 2010

888.2 Million
(ftp.isc.org/www/survey/reports/current/

Jan 2012)

2,267 Million Users

(InternetWorldStats.com, Dec 31, 2011)

(approx. 6 B mobiles and >1.5 Billion PCs)

Regional Internet Statistics 12/31/2011

Region

Internet Population Asia 1020 Mil. Europe 500.0 Mil. North Am. 273.1 Mil. LATAM/C 235.6 Mil. Mid-East 77.0 Mil. Oceania/Aus 23.9 Mil. Africa 140.0 Mil. TOTAL 2,267.0 Mil.

% penetration 26.2 % 61.3 % 78.6 % 39.5 % 35.6 % 67.5 % 13.5 % 32.7 %

5

Bob Kahns Open Architecture Idea

Each distinct network would have to stand on its own and no internal changes could be required to any such network to connect it to the Internet. Communications would be on a best effort basis. If a packet didn't make it to the final destination, it would shortly be retransmitted from the source.

Black boxes would be used to connect the networks; these would later be called gateways and routers. There would be no information retained by the gateways about the individual flows of packets passing through them, thereby keeping them simple and avoiding complicated adaptation and recovery from various failure modes.
There would be no global control at the operations level.

Other key issues

Global addressing Gateway functions to allow it to forward packets including routing using, handling interfaces, breaking packets into smaller pieces if necessary, etc. Algorithms to recover from lost packets.

Providing "pipelining" so that multiple packets could be en route

End-end checksums, reassembly of packets from fragments and detection of duplicates, if any. Techniques for host to host flow control. Interfacing with the various operating systems

There were also other secondary concerns, such as implementation efficiency, internetwork performance

Important Design Concepts

No particular application drives Internet design! Layering (design factorization) ARPANET layering (circa 1970) Open Systems Interconnection (circa 1978)

Internet Protocol specification does not mention routing

IP Packets dont know what they carry or how they are carried

Non-national IP address structure

Best-efforts performance target + end-to-end principle End/End Security (IPSEC)
8

Role of Openness
Open Source - Linux, Chrome, Chrome OS, Android, show source HTML

Open access - build a piece of Internet and connect it

No Intellectual Property claims for TCP/IP

Open Standards (e.g. IETF, etc.)

Commercial drivers (led to massive infrastructure investment)

Broadband open access (wholesale issue, net neutrality)

Recent Changes to Internet

IPv6 - in parallel with IPv4 [IPv6 World Launch 6/6/2012] Internationalized Domain Names New gTLDs (2000 applications) Domain Name System Security (DNSSEC) Digitally-Signed Address Registration (RPKI) Sensor Networks Smart Grid Mobile Devices
10

Internationalized Top Level Domains

11

Proposed Internationalized Domain Names

12

Security Problems
Spam Viruses/Worms/Trojan Horses
Infected websites Infected thumb drives Infected CD-ROMs, DVDs, etc.

DOS and DDOS attacks Social Engineering Poor passwords (re-usable especially) Phishing, Pharming (DNS compromise) IP address poaching Spectacular human error (configuration esp.)
13

Origins of Security Weaknesses

Weak operating systems Nave Browsers (running with too much privilege) Poor access control practices Improper configuration of hosts and clients Compromised lap/desktops + servers
Leading to BOTNETS Hackers, Organized Crime, State-sponsored cyberwarfare

14

Privacy Problems
Lax user behaviors (social networking, unsafe configuration) Weak protection of personal data by business, government
Lost laptops, thumb drives Weak security configuration of servers Poor access controls Bad business practices (sharing personal information)

Invasive devices
Mobiles with cameras and sound recording capacity

GPS tracking
RFID tracking Closed Circuit TV surveillance (traffic, stores, public places)

15

Cloud Collaboration
Multiple Data Centers (replication, responsiveness) Dynamic capacity sharing Data sharing (latest copy is always in the cloud Video and audio conferencing with data sharing Inter-Cloud interactions (new research area!)
How to refer to other clouds? How to refer to data in other clouds? How to make data references persistent (unlike URLs)? How to protect Clouds from various forms of attack (inside, outside)?

How to establish an access control regime (inside, between clouds)?

What semantics can we rely on with inter-cloud data exchange? What notion of object would be useful for inter-cloud exchange?

16

Internet Research Problems

Security at all levels Internet Erlang formulas QOS debates (smart routers?) Internationalized Domain Names (ccTLDs & GTLDs) Mobility, persistence (processes, connections, references) Multihoming

Multipath routing
Broadcast utilization

Distributed Algorithms
Presence (multi-level)

Mesh and Sensor networks

Virtualization (net, storage, processing)
17

Internet Research Problems - 2

Authentication, Identity, Authorization Governance - Law Enforcement

Multi-core Processor Algorithms

- Policy Development Delay and Disruption Tolerance - Homologation

Integration of Applications (e.g. drag/drop gadgets in calendar)

Intellectual Property Protection (tracking rights, enforcement) Role of Layering

- Facilitation of electronic commerce - Privacy and confidentiality

18

Internet Research Problems - 3

Mobile operation Dynamic joining (new IP address?, Authentication?) Dynamic Routing (Dynamic Topology) Persistent connection (ID at TCP/UDP/RTP layer?)

Self-organization (security risks?)

New Routing Concepts

- OPENFLOW
- Content-directed Routing

19

Internet Research Problems - 4

Performance Latency, Throughput (cf: Greg Chessons work) Resilience (AS partitioning) Rate management vs CWND flow control Route convergence Addressing (Identification) What objects should be addressable? New bindings of IP to Identifiers (new DNS?)
20

Policy Considerations
Broadband and wireless infrastructure Non-discriminatory access (net neutrality)

Common Carriage style safe harbors regarding content

Permissionless innovation Cloud computing efficiencies Multi-lateral ICANN Affirmations Continuation of IGF Keeping the ITU focused on traditional telecommunication and standards Coping with TLD expansion

21

Challenges of the Digital Age

Intellectual property treatment
Digital material is easy to copy and distribute

Semantic Web
Complex objects that can only be rendered via computer
3D interactive objects
Complex spreadsheets Interactive environments

BIT ROT!
Preserving interpretive programs (Windows 3000 and PPT 1997) And the operating systems that run them And the hardware that run the operating systems

For thousands of years!!

22

Internet-enabled Devices

23

24

The Data Will Come to Us!

Sensor networks everywhere!
Seismic, oceanographic, atmospheric, satellite

Smart buildings, instrumented highways

Video sensors (traffic, security)

The Smart Grid

Power metering applications
Hybrid and Plug-In vehicles Injecting power into the Grid Tracking usage for better lifestyle decisions Office and building efficiencies

25

Sergey Brins Challenge

Re-Invent the Internet so that: No central registrations/assignments required Essentially impossible to shut down Anonymity can be protected Freedom of expression can be protected Protection from online harms is possible The system is sustainable WHAT WOULD THAT DESIGN LOOK LIKE?
26

InterPlaNetary Internet

27

28

MARS RECONAISSANCE ORBITER, MARS EXPRESS, PHOENIX, MARS ROVERS, MARS SCIENCE LABORATORY

29

Interplanetary Internet:InterPlaNet (IPN)

Planetary internets

Interplanetary Gateways Interplanetary Long-Haul Architecture

Licklider Transport Protocol (LTP) Bundle Protocol (RFC 5050) Delayed Binding of Identifiers Email-like behavior

(RFC 4838)

TDRSS and NASA in-space routing Delay and Disruption Tolerant Protocols
Tactical Mobile applications (DARPA) Civilian Mobile applications (SameNet!) Deep Impact Testing October 2008 Space Station Testing July 2009 (and continuing) EPOXI Testing October 2009 (aborted) EPOXI visits Hartley-2 November 2010

30

End-to-end information flow across the solar system Layered architecture for evolvability and interoperability IP-like protocol suite tailored to operate over long round trip light times Integrated communications and navigation services

31

Next Stop: Alpha Centauri (DARPA 100YSS)

32