Transcript The end-to-end arguments vs. the brave new world

W02 / COSC 6590A / Advanced Computer Networks
Paper Presentation
Rethinking the Design of Internet:
The End-to-end Arguments vs. the Brave New World
Marjory S. Blumenthal
Computer Science and Telecoms Board, NRC
David D. Clark
MIT Lab for Computer Science
Appeared in ACM Transactions on Internet Technology, Vol. 1, No.1 Aug.2001, p70-109
Presented by Yanling Wang
Computer Science, York University, Toronto
2002-11-28
1
Outline
•
•
•
•
•
•
•
•
Communications & Internet histories
End-to-end arguments
New requirements in today’s communication
Identify a range of possible solutions that might be used to
meet these requirements
Look at the implications for the rights and responsibilities of
the various parties that comprise the Internet
Describe the range of emerging players
Conclusions
Picture of future Internet
2002-11-28
2
History of Communications
• 1500's Chief noblemen (e.g. the Pope) possessed more than 100 messengers
• 1544 Holey Roman Empire granted monopoly to the Thurn and Tassis family ->
•
•
•
•
•
•
•
•
Deutche Bundespost
1840's Telegram The first digital method of electronic communication
1876 Bell telephone developed by accident! AT&T owner of the patent.
1896 Telephone dial invented by Almon Strowger, a Kansas City undertaker.
Feared he was loosing business to a competitor whose wife happened to be a local
switchboard operator. First place to use it was La Porte, Indiana. Also developed
the 'step' exchange still in use.
1890's Loading coil developed to allow 'long-distance' calls.
1920's Telephone companies started using Strowgers technology as patent ran
out.1920's AT&T developed way of putting many calls on one line ('analogue
carrier' technology). Continued to be develop up until 1970's
1962 First communications satellite launched
1970's Fiber optic technology
1980's Digital technologies
2002-11-28
3
History of Internet
(INTERnational NETwork / INTERconnected NETwork)
1957 Sputnik Launched
1969 ARPnet: in US
1972 ARPnet: Report Generated
1973 ARPnet: Norway & England
1974 Transmission Protocol
1983 TCP/IP
1984 MILNET & DNS
1985 1st Internet Domain
1986 Online Forum & Firewall
1990 ARPnet off, Internet On
1991 Info. Retrieved from Server
1992 Hypertext  WWW
1993 Web Browsers  Surfing
1994 Size Double/Year
2002-11-28
The figure was copied from the Internet.
1999 ISPs: Internet Free Access
4
Internet Design Principles:
End-to-end Arguments
• The function in question can completely and correctly be
implemented only with the knowledge and help of the
application standing at the endpoints of the communications
systems. Therefore, providing that questioned function as a
feature of the communications systems itself is not possible.
– This suggests that specific application level functions usually can not,
and preferably should not, be built into the lower levels of the systemthe core of the network.
• Result in:
– Function migrates to the end node.
– Network should be “as transparent as technology permits”.
2002-11-28
5
A simple View of Internet
User
User
Router
Router
User
Router
User
Router
User
Router
Router
The Internet
User
User
Router
User
2002-11-28
Router
User
6
Advantages of the End-to-end Arguments
• User empowerment
–
Run what you please
–
A network to glue computers together
• Flexibility in the face of unknown applications
• Lower cost in core of network
– Eliminate special features
– Rely on edge-node equipment
• More robust applications
– No unexpected failures of third party nodes
• Moving application specific functions out of the core of the network
– Complexity of the core network reduced
– Generality in the network increases the chances that a new application can be added without
having to change the core of the network
– Applications do not have to depend on the successful implementation and operation of
application specific services in the network
2002-11-28
7
Examples of Emerging Requirements
in Today’s Internet
• Operation in an untrustworthy world
– Imply more mechanism in the center of the network to enforce “good” behavior
• More demanding applications (streaming audio & video, etc.)
– More sophisticated internet service is needed to assure each data stream a
specified throughput
• ISP services differentiation
– ISP-specific intermediate servers
• The rise of third-party involvement
• Less sophisticated users
What do these requirements really imply?
– World is becoming more complex
– What is needed is a set of principles that interoperate with each other, some
built on the end-to-end model, and some on a new model of network-centered
function.
2002-11-28
8
What About in the Brave New World?
• The End-to-end model does not empower:
– ISPs: want to sell services, add value, and make money.
New network services, protection, control of
applications/content, accounting
– Rights holders
– Governments: control of content, taxation, consumer
protection, law enforcement
– Employers
• It empowers:
– Only certain application makers
2002-11-28
9
End-to-end Argument Functions
at Network level - “in” the Network
• Adding functions to the core of the network
– Firewalls
– Traffic filters
– Network address translation elements
• Design issues in adding mechanism to the core of the
network
– Imposing a control element into the path of communication
– Revealing or hiding the content of message
– Labels on information
2002-11-28
10
End-to-end Argument Functions
at Application Level - “on” the Network
• Application-level services being employed to augment or
modify application behavior
– Anonymizing message forwarders: It is critical that the user construct
the route, the ISP, or any other third party should not be able to
determine the path directly. Third party removes the possible
identification in the messages.
– Helpful content filtering
– Content caches
• More complex application design-using trusted third parties
– Public-key certificate: user can create a public key and give it to others,
to enable communication with that user in a protected manner. Third
party issues a public key certificate and manages the stock of such
certificates.
2002-11-28
11
A Complex View of the Internet
User
User
User
Little
ISP
Backbone
ISP
Backbone
ISP
User
Campus
User
The Internet
Corp
Backbone
ISP
User
Little
ISP
User
User
User
2002-11-28
12
Addressing Where We Are
• The rise of the new players
–
–
–
–
Governments
ISPs
Industry consolidation
Institutional providers of Internet services: corporations, schools, and non-profit
organizations
– International nature of the Internet
• The erosion of trust
– May be most fundamental for transforming the Internet
– Use trusted third parties
– Both end-points and third parties may wish to interpose intermediate elements into a
communication to achieve their objectives of verification and control
¿ But, are third parties actually trustworthy? Or are end-points talking to the third party
they think they are?
• Rights and responsibilities
– The end-to-end design of the Internet gives the user considerable power in determine
what applications he/she chooses to use
– Today the Internet places few limits on what groups of consenting end-nodes do in
communicating across the network
¿ How can the desire for privacy and anonymity be balanced against the need for
accountability, given the freedom of action that the end-to-end arguments imply?
2002-11-28
13
Moving Forward
• Labels
– A comprise between autonomy and visibility of action
• Distinction between private and public communication
– Accept that private communication is not restricted
– Focus on communication to the public
• New principles for application design
– Do not force an end-node implementation
– Allow the user to select an alternative
– A more sophisticated form of empowerment
• Tolerance for experimentation
2002-11-28
14
Conclusions
• Elements that implement functions that are invisible or hostile to end-to•
•
•
•
•
•
end application, in general, have to be “in” the network
Multiple forces seem to promote change within the Internet that may be
inconsistent with the end-to-end arguments
Less work by consumers may imply less control over what they can do on
the Internet and who can observe what they do
Lost of trust
Trust can be supported by systematic labeling of content
It is no longer the single creative person in the garage, but the startup with
tens of millions of dollars in backing that is doing the group innovation
Commercial investment will go elsewhere, in support of short-term
opportunities better met by solutions that are not end-to-end, but based on
application-specific servers and services “inside” the network
2002-11-28
15
Picture of Future Internet
• It is possible that we will see, not a sudden change in
•
•
the spirit of the Internet, but a slow ossification of the
form and function.
In time some new network will appear, perhaps as an
overlay on the Internet, which attempts to reintroduce a context for unfettered innovation
It is premature to predict the final form. What we can
do is to push in ways that tend toward certain
outcomes
2002-11-28
16
The End
2002-11-28
17