IP Addresses and Subnetting


IP (Internet Protocol) addresses are 32-bit numbers that identify a specific device on an internetwork. These addresses are broken down into four octets (a synonym for byte) and are written with periods (dots) separating the octets. For example, an IP address for a particular machine would be written as

206.171.100.1

IP adddresses are broken down into 3 network classes: Class A, Class B, Class C. A class A IP network uses the first octet (8 bits) for the network number, and 24-bits (the last three octets) for the host machine number. A class A address therefore can represent up to 254 different networks, each having up to 16,777,216 host computers. A Class B address uses 16-bits for the network address and 16-bits for the host address, giving 65,533 networks, each with 65,533 hosts. A Class C address can have 16,777,214 network addresses, with each network having 254 hosts. Note that a host can be a server, a desktop computer, a router, a terminal server, or some other device. Each of these devices requires a unique IP address if it is to communicate over a TCP/IP network.

In order to identify the number of bits reserved for the network number, the concept of the netmask was devised. A netmask for a Class C address would be written as

255.255.255.0

indicating that the last 8-bits of the address are reserved for host addresses.

The concept of classes has recently given way to another form of specifying the number of bits reserved for network addresses. This new terminology refers to an IP address followed by the number of bits reserved for the network number. For example

207.215.121.0/24

is now the preferred method of indicating what was formerly referred to as a Class C address. Because the slash character separates the network number from the number of bits in the subnet mask, it is referred to as a "slash 24". This method combines the network address and the subnet mask in one. Newer routers can recognize this notation.

Classless Internet Domain Routing, or CIDR (pronounced like "cider"), is an attempt by the IETF to address the explosion in the proliferation of IP addresses on the Internet. Most people have been led to believe that CIDR addressing is an attempt to conserve IP addresses. The actual benefit is to allow ISPs to aggregate the number of routes in their router's routing tables. While CIDR addressing does allow ISPs to assign addresses based on the actual size of a customer's network, it actually wastes some otherwise usable addresses. Howvever, this loss of address space is offset by the need to aggregate routes, since this allows routers to operate more efficiently, resulting in the faster transmission of traffic on the Internet.


WAN Page Ray's Home Page E-Mail Me

Creation Date: Saturday, October 19, 1996
Last Modified: Saturday, November 30, 1996
Copyright © Ray Smith, 1996