Understanding the Difference in IP Address Classes

Understanding the Difference in IP Address Classes

IP addressing is the logical addressing system that allows complex, geographically distributed computer systems to exist. Thanks to the IP addressing schemes made possible by the TCP/IP protocol suite, advanced devices like routers can effectively and dynamically move packets of data from a source network to a destination network located anywhere in the world.


These days, many small networks – such as those in homes or offices with fewer than a dozen employees – can be provisioned with IP addresses automatically. In most cases, however, it is necessary for a network administrator to develop a cohesive IP addressing scheme based upon the consistent use of one of the three main IP address classes.

In truth, even the smallest dynamic network will be prone to performance issues. In some cases, these issues can even present security vulnerabilities that might expose sensitive data to attack. That being the case, it’s important that a well-rounded network architect be familiar with the IP address classes and how to deploy appropriate addressing in different types of networks.

Consistent, logical IP addressing is also crucial for another reason. As cloud technology has matured, more and more networks have virtualized their capabilities. To make effective use of virtual machines, network engineers need to have a firm grasp of topics like IP addressing and subnetting. A lapse in these areas could render network assets inaccessible. Mobile access to networks often depends on strong principles of virtualization.

The Five Classes of IP Address and How They are Used

Every IP address is made up of 32-bit binary numbers. These addresses need to effectively identify both the network and each one of the individual hosts on that network. A “host” could be an end user terminal, a server, peripherals such as printers, or a whole variety of other devices.

For most of the history of the Internet, three main address classes have been used:

  • Class A: Class A addresses are assigned to large global enterprises such as governments, nonprofits, and the top commercial entities. In such an address, the high-order bit is zero, while the next seven bits – that is, the first octet – serve as the network identifier. The final 24 bits are devoted to host IDs, allowing 126 Class A networks with 16,777,214 hosts each.

  • Class B: With a total capacity of 16,384 networks with 65,534 hosts each, Class B networks represent a middle ground suitable for large and mid-sized entities. The two high-order bits are always set to 1 and 0 respectively. The next 14 bits (that is, through the first two octets) are devoted to the network ID; the next 16 bits are for the host ID.

  • Class C: Class C networks provide the smallest number of individual hosts: 2,097,152 networks can have 254 hosts each. The three high-order bits in a Class C network must be set to binary 1, 1, 0. The next 21 bits – through to the end of the first three octets – are part of the network ID, while the last eight bits identify individual hosts.

Although the number of possible IP addresses is extremely high, there is still a limit: IPv4 provides for 4,294,967,296 total addresses, of which about 3,706,452,992 were in use by 2010. To combat the ongoing concerns of IPv4 address exhaustion, researchers have been working on IPv6 for decades. IPv6 will succeed IPv4 and provide 340 undecillion addresses – that is, 340 trillion trillion trillion.

Individual network administrators have no influence over whether their enterprise is Class A, B, or C, but will be responsible for ensuring addressing is carried out in an efficient and effective way. In unusual cases, some computer scientists might find themselves dealing with Class D or Class E addresses. These are reserved for special purposes.

  • Class D: Class D addresses are only used for IP multicasting. The four highest-order bits in these addresses must reflect the binary values 1, 1, 1, 0. The other bits are used for the address that impacted hosts will recognize. Microsoft supports the use of these addresses, allowing applications to multicast to compatible hosts within a given network.

  • Class E: This experimental address class is reserved for future applications.
  • Because of the dynamic and limited nature of current IP addressing, computer scientists need to maintain a clear awareness of their applications and limitations. For the foreseeable future, it will be necessary to maintain fluency in both IPv4 and IPv6 IP addressing conventions. The robust nature of IPv6 may change the standards by which addresses are assigned, but it will still be up to technical experts to ensure resources are networked in the most efficient and effective way. Thanks to the proliferation of devices that use IP addresses, the interrelationship between the different address classes will remain one of the most fundamental knowledge areas in the worlds of information technology and computer science.

    New Jersey Institute of Technology offers an online master in computer science program so students can learn the latest concepts and techniques in Computer Science, and develop the skills to take the world in imaginative new directions.