handley future of internet

8/14/2019 Handley Future of Internet

1/21


2/21

A Global Network

English34%

Dutch2%

French4%

Arabic

2%

Malay2%

Other

4%

German7%

Italian3%

Scandinavianlanguages

2%

Russian1%

Polish1%

Spanish9%

Korean4%

Chinese14%

Japanese8%

Portuguese

3%

0

100

200

300

400

500600

700

800

900

1000

Dec-96

Dec-97

Dec-98

Dec-99

Dec-00

Dec-01

Dec-02

Dec-03

Users(Millions)

Languagesof Internet

Users

People on the Net


3/21

The net is a success!

The problem:

In almost every way, the Internet only just works!


4/21

The net only just works?

Its always been this way:1975-1981:

TCP/IP split as a reaction to the limitations of NCP.

1982:

DNS as a reaction to the net getting too large for hosts.txt files.

1980s:

EGP, RIP, OSPF as reactions to scaling problems with earlier routingprotocols.

1988:

TCP congestion control in response to congestion collapse.1989:

BGP as a reaction to the need for policy routing in NSFnet.


5/21

Changing the net.

1st Jan 1983.

Flag day.

ARPAnet switched from NCP to TCP/IP.

About 400 machines need to switch.

QuickTime and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Sweden Changeover to

Right Hand Traffic 1967

As the net got bigger, it got a lotharder to

change.


6/21

Before web...

Prior to the 1990s the Internetwas primarily academic and scientific.

Common goals.

Low cost of failure.

Then came the web, and commercialization of the Internet.

Exponential growth.

Financial costs of failure.

ISPs struggling to keep ahead of demand.

Huge innovation in applications.

0

50

100

150

200

250

300

350

400

Aug-

83

Number of computers

on the Internet

Number of computers

on the Internet


7/21

Development Cycle

QuickTime and a TIFF (Uncompressed) decompressor are needed to see this picture.

QuickTime and aTIFF (Uncompressed) decompressor


We need this new feature tokeep our network functioning

Heres a solution.

Let us know how it works.



8/21

Hill Climbing

Internet

PossibleFutureInternet

Youare

hereYouar

e

here


9/21

An Example:

Running out of addresses...

The current version of the Internet Protocol (IPv4) uses 32 bit

addresses.

Not allocated very efficiently.

MIT has more addresses than China.MIT + Interop trade show + Halliburton

IPv6 is supposed to replace IPv4.

128 bit addresses. We dont need to be smart in address allocation.

How do we persuade people to switch?


10/21

Network Address Translators

Scarcity of addresses has made addresses expensive.

tiered

pricing

10.0.0.2

10.0.0.3

128.16.0.1

From 128.16.0.1,

TCP port 345

From 128.16.0.1,

TCP port 678

From TCP

port 222

From TCP

port 222

PublicInternet

NAT

NATs map one external address to multiple private internaladdresses, by rewriting TCP or UDP port numbers in flight.


11/21

Network Address Translation

Introduces asymmetry: Cant receive an incoming connection.

Hard to refer to other connections:

Eg. SIP signalling causes the phone to ring.

On answer, set up the voice channel.

Application-level gateways get embedded in NATs.

Cant change the ends until you change the middle. Middle wont change til ends demonstrate a need.

It should be easy to deploy new applications!


12/21

Digital Convergence:One Network Connecting Everyone

Telephone

Network

Telephone

Network

DataNetwork

DataNetwork

TV

Network

TV

Network

Mobile

Telephone

Mobile

Telephone

Global

Internet

Global

Internet


13/21

QuickTime and aTIFF (LZW) decompressor


Normal dailytraffic

The Reality:

Internet Worms, Viruses, and Denial-of-Service Attacks

5:30am GMT

24th Jan 2003

Slammer Worm


14/21

The sky is falling!!!

No.

But were accumulating problems faster

than theyre being fixed.

There has been no significant

architectural change to the network corein a decade.

The consequences of failure are

growing.



15/21


16/21

Key Challenge

Is it possible to change the Internetarchitecture in a planned way,

so as to achieve long-term goals?(or is it only possible to patch the pieces repeatedly until it gets too expensive and

unreliable, and eventually something better comes along and replaces it?)


17/21

Evolving the Internet Architecture:

Changing the Engines in Mid-Flight


18/21


19/21

What can be done?

Basic long-term research. Industry players cant afford the long view.

NSFs FIND programme.

Close the loop between research and real-worldexperiments.

GENI, XORP open source router, Route Views/RIPE routing databases.

Coordination, coordination, coordination.

Need to find a way to involve equipment vendors, ISPs, and researchers.

CRN: Communications Research Network,

IRTF Internet Congestion Control Research Group.


20/21

Our Research Goal

Make the Internet worthy of the trust that is being placed

in it.


21/21

The End(of the beginning)

ARPAnet, 1974

handley future of internet

Documents