All told, there are over 4.7 billion possible host addresses. Unfortunately, the four-octet structure places heavy restrictions on the distribution of these addresses. Every time a Class A address is assigned to an organization, almost 17 million host addresses go with it. If all 126 of the Class A addresses are assigned, then over 3 billion of the 4.7 billion possible addresses are gone. If all of the 16,000 Class B addresses are assigned, then another billion host addresses are gone as well. Whether or not all the workstation addresses are actually put to use or not is irrelevant; they have been assigned to a specific network and cannot be used by another organization.
Class C addresses represent the biggest problem however, for two reasons. First, there are less of them than with the other network classes (only about 500 million possible node addresses are available from all of the Class C networks combined). Second, they are the most popular, since they reflect the size of the majority of the LANs. However, every time you assign a Class C address to a network segment, you take 254 possible node addresses with it. Organizations who have three segments but only have sixty devices are wasting over 700 possible addresses (3 segments x 254 node addresses = 762 addresses - 60 active nodes = 702 inactive addresses).
To some readers, the logic for having different "classes" of addresses may seem vague at best. With the current design, there are only 2,113,662 possible networks. If all networks used the first 24 bits (without using "class bits") to identify the segment, there would be a possible 16,777,124 networks, with 254 nodes on each of them.
Remember however that TCP/IP networks are inherently router-based. It requires much less overhead on the part of nodes and routers to remember a few networks than to remember thousands of them. Having to process 16 million networks would quickly overwhelm the router databases, and network traffic would slow down tremendously. Having network classes allows routers to deal with large networks easily, and therefore performance can be maintained.
Remember also that the original architecture of the Internet consisted mostly of large networks connecting to each other. It was easy to give one huge address block to milnet (a network of military systems) and another to NSFnet (the National Science Foundation's network). By doing this, routers only had to remember one other router's address in order to pass data to millions of hosts.
A new version of IP has been developed that will overcome most of these limitations. However, it will be several years before IPng ("IP next generation", or IP v6) will be implemented on enough commercially available equipment for you to take advantage of it on an enterprise-wide basis.
For more information about 32-bit addressing, refer to section B.3.1 Binary Addresses, Octets and Network Classes. For more information on address availability, refer to section B.3.3 Internet-Legal versus Private Addressing. For information on how to allocate addresses, refer to section B.3.4 Sub-Net Masks.
Copyright © 1997, Unoverica Corporation. All rights reserved. Unauthorized use prohibited.
Send comments to docs@unoverica.com