The bedrock of the current cyber infrastructure is the Internet. In the fol-
lowing sections we present a brief overview of the Internet’s architecture
and history, focusing in particular on those aspects that are relevant to the
later discussion of the I-2 infrastructure. A reader seeking a more complete
discussion of the Internet is referred to [Comer 2006, Gralla 2006, Leiner et al.
1997]. Both the architecture of the Internet as well as the strategic decisions
that its architects made to facilitate its growth embody important lessons
about building successful distributed global infrastructures. The discussion
in this chapter is a preamble to the discussion in Chapters 7 and 8 in which
we discuss the guidelines for architecting the emerging distributed global
A Simple Analogy
The Internet is a globally distributed infrastructure that provides electronic
real- time data transport service. It bears some resemblance to the globally
distributed postal service infrastructure. Whereas the postal service trans-
ports packages, the Internet transports digital packets called IP datagrams.
Just as an entity has to have a postal address to be able to receive a postal
shipment, an end node in the Internet has to have an IP address in order to
be able to receive an IP datagram. The global postal service network is not
controlled or operated by a single agency. Instead, it comprises a network of
independently operated postal service infrastructures—one in each coun-
try—that cooperate to provide seamless global service. The Internet too is
not owned or managed by a single entity. It is a global infrastructure that
relies on the cooperation of several independently managed networks. A
postal shipment typically goes through a multihop journey visiting several
intermediate processing and distribution centers before reaching the desti-
nation. An IP datagram is also typically routed through several intermediate
routers in its journey.
Although the Internet has several idiosyncrasies that distinguish it from
the simple postal service—and we discuss the details of such idiosyncratic
features below—the simple analogy, described above, might serve as a use-
ful backdrop for the discussion in the remainder of this chapter.
50 Design and Construction of an RFID-Enabled Infrastructure
Architecture of the Internet
The architecture of the Internet is best described by tracking a hypothetical
online transaction. Accordingly, we consider a user who seeks to access a
remote web page from a computer at his house. We will assume that he has
multiple computers that connect to the Internet through a single gateway in
his house. A gateway is a connection point between the user’s home network
and the external Internet. All communications between the Internet and the
user’s home network ow through the gateway. The user starts by specify-
ing a web address—say http://www.purdue.edu/ index.html—to his browser
instructing it to retrieve the web page at the specied address. His request
triggers a sequence of communications, which are shown ordered alphabeti-
cally in Figure 3.1. The events triggered by his request are outlined below
and elaborated further in the sections that follow.
1. When the user’s computer connects to the home network the gate-
way assigns an internal IP address to the user’s computer, using
DHCP—Dynamic Host Conguration Protocol. (For simplicity, we have
assumed that the gateway also serves as a DHCP Server for his home
2. The web address http:www.purdue.edu/index.html is forwarded by the
browser, through the gateway, to a DNS Server at the local Internet
Service Provider (ISP).
3. After a series of delegated requests, the local ISP’s DNS Server
returns the IP address of the web server corresponding to the address
Coarse- grained illustration of the Internet architecture.