Understanding IPv4 address structure and classes
Understand IPv4 prefixes through one /24 example, how it splits into /25 blocks, and why modern CIDR is more useful than guessing an address class.
You are setting up an office network and see 192.0.2.0/24 in an example. Someone calls it a “Class C”. Which part actually tells you its size? Today, the useful answer is the prefix length: /24. The old class label is historical context, not a substitute for reading the configuration.
The addresses below come from a range reserved for documentation. They illustrate the arithmetic, not a public network you should deploy.
Four numbers represent one 32-bit address
IPv4 writes 32 binary digits as four decimal numbers separated by dots. Each group represents eight bits and ranges from 0 to 255. The dots separate the groups; they do not, on their own, tell you where the network prefix ends.
A prefix length provides that boundary. In 192.0.2.0/24, the first 24 bits are fixed and eight remain for addresses within the block. Eight variable bits give 256 combinations, from 192.0.2.0 to 192.0.2.255. The equivalent mask is 255.255.255.0.
A smaller prefix number means a larger block
The notation can initially feel backwards. A longer fixed prefix leaves fewer variable bits. A /25 contains 128 addresses; a /24 contains 256; a /22 contains 1,024.
Our example /24 can split into two aligned halves: 192.0.2.0/25 and 192.0.2.128/25. Each half has its own prefix. The division changes how addresses are grouped, not the length of an individual IPv4 address.
On a conventional broadcast subnet, the network and broadcast addresses are not assigned as ordinary host addresses, leaving 254 in a /24. Do not turn “subtract two” into a universal rule: point-to-point links can use a /31, where both addresses identify endpoints.
What were Classes A, B and C?
Early classful addressing used fixed divisions. Class A used an eight-bit network prefix, Class B sixteen bits and Class C twenty-four. The first bits of an address identified its class. This made the available block sizes coarse: an organisation could need more than one size supplied but far less than the next.
The historical Class D range is for multicast, sending traffic to a group. The historical Class E range is reserved space, not another ordinary pool for company hosts. The IANA address-space registry records special ranges and their status.
Why modern networks use CIDR
Classless Inter-Domain Routing, or CIDR, makes the prefix explicit instead of inferring a fixed size from the address class. It lets planners use different block sizes and lets routing aggregate compatible, aligned blocks. The CIDR specification explains the address-assignment and aggregation model.
Aggregation is not a licence to combine unrelated addresses into an arbitrary larger route. The blocks must fit the covering prefix, and the routing arrangement must actually support reaching its destinations. A neat spreadsheet alone does not establish a working path.
Three questions before changing a network
- What is the configured prefix? Read the mask or prefix length, rather than guessing from the first number.
- Which addresses are assigned, reserved or still in use? Confirm the running network as well as the records.
- What else depends on this range? Routes, firewall rules, DNS and remote connections may all need attention.
The introduction to address management explains how to keep those records useful. A prefix describes a technical grouping. It does not by itself identify a legal owner or give a registry political authority over the networks using it.
Why the arithmetic does not settle the bigger argument
CIDR uses the existing 32-bit space more flexibly; it does not make it infinite. But neither does the finite number, by itself, tell an operator what addresses are available or which investment is worthwhile.
In Note 45, Lu Heng challenges the use of address scarcity as a justification for discretionary institutional control. Knowing what an address and a prefix actually do helps readers examine that claim without confusing a technical coordination task with authority over everyone who depends on it.