Network, Computer and Programming Resources

Network - IP Address Classes

The following classes are defined for IP addresses: n Class A is used for very large networks (networks with a large number of nodes). This class uses 7 bits for Net and 24 bits for Host. The high-order bit is 0 in such an address. There are 128 class A networks possible. The now-defunct ARPANET, which had a network address of 10, is an example of a network in this class. n Class B is used for medium-size networks, such as networks that span a large college campus. This class uses 14 bits for Net and 16 bits for Host. The two high-order bits are set to 10. This address class is also popular for local-area networks (LANs), particularly if they use subnetting. n Class C is used for small networks (those with no more than 255 nodes). This class allocates 21 bits for Net and only 8 bits for Host. The three highorder bits are 110. 488 IP Address Classes n Class D allocates 28 bits for a special multicast address, which is an address in which a group of targets are speci- fied. The first 4 bits of such an address are always 1110. n Class E is a reserved address class. Addresses in this class are for experimental use, and cannot be guaranteed to be unique. The first 4 bits of this type of address are always 1111. IP ADDRESS BREAKDOWN In summary, address classes are distinguished by the high-order bits: 0 for class A, 10 for class B, 110 for class C, 1110 for class D, and 1111 for class E. Two particular Net addresses-0 and those with all 1s-are reserved. Net address 0 is reserved for the originating entity (network or host), and address 255 is used for broadcasts. Figure "IP address breakdown" shows how the bits are allocated for the different address classes. The use of subnetting provides additional flexibility in addressing. A subnet is a portion of a network or an internetwork that can be viewed from the outside as a single element. An IP address that uses subnetting has three types of information: network, subnet, and host. Subnets are identified by combining an address with a mask, which is a bit pattern that cancels out unwanted bits, so that only the bits of interest remain. PRIMARY SOURCE RFC 1349