The Internet has been receiving a lot of attention in the media lately, but it remains a mystery to many. It has been described as everything from a giant bulletin board to an Information Super Highway. While these phrases invoke some interesting imagery, they don't really do much to explain the real nuts and bolts of the Internet. This chapter will attempt to dispel some of the confusion.
The Internet is really just a collection of computers and computer networks all speaking the Internet Protocol (IP). There is no single organization that owns or controls the Internet; it is a cooperative effort composed of many universities, government agencies, businesses, and more.
The Internet grew out of the government funded communications network ARPANET. ARPANET was created by DARPA, the Defense Advanced Research Projects Administration, in the mid 1970's. It tied together various government agencies, universities, and government contractors. In the early 1980's ARPANET began using TCP/IP (Transmission Control Protocol / Internet Protocol). TCP/IP gained even wider acceptance when it became a standard part of BSD UNIX, a computer operating system in heavy use in universities at the time. TCP/IP is still the underlying network protocol that drives the Internet today.
The National Science Foundation joined the act in 1986 and founded NSFnet. They funded the Internet backbone network, a collection of high speed communication lines that connects regional networks all over the country. They helped many universities create their own regional networks and connect those networks to the Internet. The Internet began to grow at a staggering 15 percent each month.
Over time, the number of businesses connecting began to outpace the number of universities and government agencies. In fact, the NSF funding of the backbone is going away, and are moving to a commercially funded Internet backbone. A connection to the Internet is fast becoming a standard business tool, as necessary as a phone and fax machine.
The Internet has grown considerably since its humble beginnings. Today, over 30 million people use the Internet. Over four million computers are connected to it. The Internet continues to grow at an amazing rate. In fact, if you chart projected growth in Internet users on the same graph with projected population growth, the lines intersect by the year 2001. While it is unlikely that every person in the world will have Internet access by that time, it is clear that Internet is evolving into a resource as widespread and accepted as television and the telephone.
Every computer on the Internet must be assigned a unique IP address. Much like the way a postal address is used to route letters to and from your home, an internet address is used to route information to and from your computer on the Internet. Because computers speak to each other in numbers, an IP address is really just a number. It is a 32 bit number, which means it can be an integer from 0 to 4,294,967,295. Actually, when writing an Internet number, we split it into eight bit numbers. This gives us four numbers in the range of 0 to 255 and makes the address a little easier to read and interpret. We separate the numbers with a dot (.) character to signify that they are all part of the same address.
As an example, let us take the 32 bit number 3349838082 and write it as an internet address. First, we can write out all 32 of the binary digits:
11000111101010101011000000000010
Next, we split the number into 8 bit octets:
11000111 10101010 10110000 00000010
Then, we convert each of the octets back into integers. This yields the four parts of the standard Internet address notation:
199.170.176.2
An Internet address is always comprised of two parts; the network number and the host number. These numbers can be 8, 16, or 24 bits long (1, 2, or 3 octets), thus the four part, dotted notation makes the address easier to interpret. In the above example, the first three octets are the network number, and the last octet is the host number. This is called a class C address. There are three classes of address; A, B, and C. Some examples of the different address classes follows:
With a class A address, the first octet is the network number and the last three octets are the host number. There are 127 possible class A networks with a maximum of 16777214 computers on each network.
With a class B address, the first two octets are the network number and the last two octets are the host number. There are 16382 possible class B networks with a maximum of 65534 computers on each network.
With a class C address, the first three octets are the network number and the last octet are the host number. There are 4194304 possible class C networks with a maximum of 254 computers on each network.
Numeric addresses are fine for computers, but people prefer something a little easier to remember. The Domain Name Service (DNS) is a method for mapping meaningful names to the numeric addresses. For example, you might have a file server in your legal department with the IP address 199.170.177.80. You could assign it the DNS hostname legal1. Then whenever you wish to access that server, you can use the name legal1 instead of typing the IP address.
The Domain Name Service, like the IP address itself, is hierarchical in nature. A
complete hostname (often called a fully qualified hostname) consists of several dot separated parts.
The rightmost part is called the top level domain. It is an abbreviation for the type of organization. In this example, the top level domain com signifies that it is a commercial organization. Other top level domains include edu for educational institutions, mil for military, and gov for government organizations.
Just to the left of the top level domain is the organization's domain name. In this example, the domain name glaci is the unique Internet domain name reserved for the Great Lakes Area Commercial Internet. The leftmost part of our example is the actual hostname, ren. This maps to the internet address 199.170.176.2. If a computer anywhere on the Internet attempts to connect to the fully qualified hostname ren.glaci.com, it will query the Domain Name Service with that name, receive the number 199.170.176.2 as an answer, and then connect to that IP address.
Large organizations may sometimes divide their networks into smaller, more manageable networks. This can result in an additional name component, called a
sub domain, to be written between the hostname and domain name. For example,
the University of Milwaukee, WI - USA has the domain name uwm.edu. The university is divided into multiple departments, one of which is the Computer Science Department. The department has a subdomain named csd. Within that domain is a computer named convex. Thus that computer's fully qualified hostname is convex.csd.uwm.edu.
Internet domain names are organized in a hierarchical manner, much like the organization of files and directories on a computer's hard drive. The very top of the domain tree is called the root domain, and is often represented with a single dot (.) character. This branches into the various top level domains. Each top level domain branches into various organizational domain names. The lowest level of the tree contains the actual hostnames. A partial representation of the DNS tree follows: