The new docs.trellix.com features a modernized UI and AI-powered conversational search. Content is currently available in English, with additional languages launching in early November 2026. We hope you enjoy the updated experience.

Background

Prev Next

The strategy for classless inter-domain routing (CIDR) was originally submitted to the networking community as RFC 1519 in 1993.

RFC 1519 aimed to address two of the principal networking problems of its day:

  • The exhaustion of the class B network address space.

  • The unmanageable growth of routing tables in Internet routers.

In so many words, CIDR introduced the concept of "subnetting."

Before RFC 1519, networks were "classful." That is, an explicit network mask was not used to determine the quantity of hosts a network could maintain. Instead, a class was determined based on the value of the first octet in the IP address.

Class

Value of First Octet

Assumed Network Mask

A

1 - 126

255.0.0.0

B

128 - 191

255.255.0.0

C

192 - 223

255.255.255.0

For example, an IP address of 32.x.x.x was assumed to have an 24-bit network mask (class A), an address of 155.x.x.x was assumed to have a 16-bit mask (class B), and an address of 210.x.x.x was assumed to have a 8-bit mask (class C).

According to RFC 1519, the fundamental cause of the exhaustion of the class B network address space was the lack of an appropriate class for mid-sized organizations. Specifically, a class C network allows for a maximum of 254 usable hosts (2^8) - 2, which is typically too small, whereas, a class B network provides 65534 usable hosts (2^16) - 2, which is overkill in most cases.

Note

You subtract 2 from the quantity of usable hosts to account for the network address and the broadcast address for each segment.

The overwhelming growth of routing tables is also a product of this shortcoming. Imagine your network has 1000 hosts. It would not be in the interest of your ISP to allocate your company a full class B network, yet a class C network would not suffice. Instead, the ISP would likely allocate you four class C networks. Unfortunately, the ISP would then have to include a new route for each of those four networks in its routing tables and propagate them across the Internet.

In this case, four routes would be added to the Internet routing space when a single route could have ideally achieved the same end.

CIDR deprecates the notion of network class and introduces the requirement that all networks and hosts be represented by a combination of IP address and explicit network mask. In this way, CIDR resolves both the aforementioned problems. Specifically, CIDR allows for:

  • The straightforward division of class B networks into blocks of class C networks.

  • Aggregation of those blocks into a single Internet route.