IN 1992, THE explosive growth of the World Wide
Web began. The architecture of the Internet was
commonly described as having four layers above the
physical media, each providing a distinct function:
a “link” layer providing local packet delivery over
heterogeneous physical networks, a “network” layer
providing best-effort global packet delivery across
autonomous networks all using the Internet Protocol
(IP), a “transport” layer providing communication
services such as reliable byte streams (TCP) and
datagram service (UDP), and an “application” layer.
In 1993, the last major change was made to this
classic Internet architecture;
11 since then the scale
and economics of the Internet have precluded further
changes to IP.
A lot has happened in the world
since 1993. The overwhelming success
of the Internet has created many new
uses and challenges that were not anticipated by its original architecture:
• Today, most networked devices are
• There has been an explosion of security threats.
• Most of the world’s telecommunication infrastructure and entertainment distribution has moved to the
• Cloud computing was invented to
help enterprises manage the massive
computing resources they now need.
• The IPv4 32-bit address space has
been exhausted, but IPv6 has not yet
taken over the bulk of Internet traffic.
• In a deregulated, competitive world,
network providers control costs by allocating resources dynamically, rather
than provisioning networks with static resources for peak loads.
Here is a conundrum. The Internet
is meeting these new challenges fairly
well, yet neither the IP protocol suite
nor the way experts describe the Internet have changed significantly since
1993. Figure 1 shows the headers of a
typical packet in the AT&T backbone,
giving us clear evidence that the challenges have been met by mechanisms
well outside the limits of the classic
Internet architecture. In the classic
description, the only headers between
of the Internet
A new model for describing the Internet
reflects today’s reality and the future’s needs.
BY PAMELA ZAVE AND JENNIFER REXFORD
˽ For the past 25 years, the Internet has
been evolving to meet new challenges
by composition with new networks
that were unanticipated by the original
architecture. New networks make
alternative trade-offs that are not
compatible with the general-purpose
Internet design, or maintain alternative
views of network structure.
˽ In an architecture composed of self-contained networks, each network is
a microcosm with all the basic network
mechanisms including a namespace,
routing, a forwarding protocol,
session protocols, and directories.
The mechanisms are specialized
for the network’s purposes,
membership, geographical span, and
level of abstraction.