Class B IP addresses occupy a unique and important position in the Internet Protocol version 4 address system. Designed to strike a balance between the very large networks of Class A and the numerous smaller networks of Class C, they provide an optimal address resource for medium-sized organizations. Understanding the characteristics of Class B addresses not only aids network planning but is also crucial for a deeper understanding of IP network architecture.

Class B IP addresses begin at 128.0.0.0 and end at 191.255.255.255. This range stems from a core rule governing IP address classification: the top two digits of Class B addresses are always set to "10." This characteristic becomes even clearer when converted to binary: the value of the first octet ranges from 10000000 (decimal 128) to 10111111 (decimal 191). This design allows Class B addresses to provide 16,384 unique network identifiers out of a total of approximately 4.2 billion IPv4 addresses, each of which theoretically supports up to 65,534 host devices. This size is particularly well-suited for university campuses, medium-sized businesses, or government agencies, which often need to connect thousands of devices but don't require the massive size of Class A addresses required by multinational corporations.

The standard subnet mask for Class B addresses is set to 255.255.0.0, which represents 16 consecutive "1s" followed by 16 "0s" in binary. This structure clearly divides the 32-bit IP address into two parts: the first 16 bits identify the network itself, and the last 16 bits identify specific hosts within that network. For example, in the address 172.16.35.200, according to the standard mask, "172.16" indicates the network number, while "35.200" identifies a specific device within that network. This division creates the unique advantage of Class B addresses—it provides a sufficient number of networks while ensuring that each network can accommodate sufficient devices.

However, standard subnet masks are often uneconomical in practice. Fixed sizes cannot adapt to the actual needs of different organizations and can result in significant waste of address space. Imagine an enterprise with 2,000 devices acquires a full allocation of Class B addresses. This leaves over 63,000 unused addresses. Subnetting technology was developed to address this problem. By using host bits to create smaller subnets, network administrators can fine-tune address space allocation based on actual needs. For example, using a subnet mask of 255.255.255.0 (i.e., a /24 prefix) allows a Class B network to be divided into 256 subnets, each supporting 254 devices. This flexibility significantly improves address utilization efficiency.

In practice, subnetting requires rigorous calculations. For example, suppose we have a Class B network 172.16.0.0 and wish to divide it using a mask of 255.255.255.128 (/25). In binary, this mask is 11111111.11111111.11111111.10000000. It borrows 9 bits from the original host portion, dividing the network into 512 subnets, each accommodating 126 devices. The key to calculating this is determining the network address, the range of available hosts, and the broadcast address. For the first subnet, for example, the network address is 172.16.0.0, the available address range is 172.16.0.1 to 172.16.0.126, and the broadcast address is 172.16.0.127.

With the widespread adoption of Classless Inter-Domain Routing (CIDR) technology, the traditional boundaries between Class A, B, and C are becoming increasingly blurred. CIDR allows for more flexible address allocation, freeing it from the strict constraints of traditional classes. For example, a network block with a /20 prefix, while neither conforming to the traditional Class B /16 nor the Class C /24, provides precisely 4096 addresses, precisely meeting the needs of organizations of a certain size. This evolution has made IP address allocation more efficient and significantly slowed the rate of IPv4 address depletion.

In Class B addressing, the private address block 172.16.0.0 to 172.31.255.255 plays a special role. These addresses are reserved for internal networks and are not routable on the public internet. Enterprises can freely use these addresses to build internal networks, accessing external resources simply through Network Address Translation (NAT). This mechanism significantly alleviates the pressure on public IP addresses and has become a standard practice in modern network design.

FAQ

Q: Why do Class B addresses start at 128 and end at 191?

A: This is due to the binary design of IP address classification. Class B addresses require the first two digits to be fixed as "10", and the first eight digits must be a minimum of 10000000 (128) and a maximum of 10111111 (191). This design ensures clear distinction between address classes.

Q: How many devices can a Class B network actually accommodate?

A: In theory, a standard Class B network supports 65534 devices (2^16-2). Network addresses with all 0s and broadcast addresses with all 1s need to be subtracted. The actual capacity may be reduced due to subnet division.

Q: With the popularity of CIDR, are there still pure Class B networks?

A: Although CIDR has changed the way addresses are allocated, the Class B address range (128.0.0.0-191.255.255.255) and its characteristics still exist. Network equipment and service providers still optimize and manage routes based on these ranges.

Q: How can an enterprise determine whether it needs Class B addresses or multiple Class C addresses?

A: It depends on the scale of the equipment and growth expectations. If you need more than 30,000 addresses and expect steady growth, Class B addresses are more suitable. If you require more flexible scalability, multiple Class C addresses may be more optimal. Modern solutions tend to use CIDR for precise allocation.

The design of Class B IP addresses reflects the wisdom of network architects, striking a valuable balance between scale and efficiency. While technologies like CIDR are transforming traditional network segmentation, understanding the principles of Class B addresses remains a crucial step in mastering the fundamentals of IP networking. Whether planning an enterprise network or preparing for professional certification, this knowledge provides a solid theoretical foundation and practical guidance for technicians.