Class Addressing

Class network addressing is an IP addressing method. The use of this method does not allow economical use of a limited resource of IP addresses, since it is not possible to apply different subnet masks to different subnets.

Basic Concepts [edit | edit wiki text]

Initially, IP addressing was carried out on the basis of classes: the first bits determined the network class, and according to the network class, it was possible to say how many bits were allocated for the network number and node number. In total there were 5 classes:

IP Addressing [edit | edit wiki text]

A feature of IP is a flexible addressing system. The charge for this is the availability of centralized DNS services.

The address consists of two parts - the network number and the node number in the network. The IP address of version 4 is 4 bytes long and is written as four decimal numbers separated by dots.

To determine which bytes belong to the network number and which node number there are several approaches.

One approach was the class addressing method.

It is easy to calculate that the total in the IP address space is 128 networks with 16 777 216 Class A addresses, 16384 networks with 65536 Class B addresses and 2 097 152 networks with 256 Class C addresses, as well as 268 435 456 multicast addresses and 268 435 456 reserved addresses. With the growth of the Internet, this system turned out to be inefficient and was supplemented by classless addressing (CIDR).

Classless Addressing

Classless Addressing (Eng. Classless Inter-Domain Routing , Eng. CIDR ) is an IP addressing method that allows you to flexibly manage the IP address space without using rigid class addressing frameworks. Using this method allows you to economically use a limited resource of IP addresses, since it is possible to apply different subnet masks to different subnets.

Address ranges [edit | edit wiki text]

The IP address is an array of bits. The principle of IP addressing is the allocation of a set (range, block, subnet) of IP addresses, in which some bits of the bits have fixed values, and the remaining digits run through all possible values. The block of addresses is specified by specifying the starting address and subnet mask. Classless addressing is based on the variable length of the subnet mask (English variable length subnet mask , VLSM ), while in class (traditional) addressing the length of the mask is strictly fixed by 0, 1, 2 or 3 fixed octets.

Example of subnet 192.0.2.32/27 using classless addressing:

In this example, you can see that in the subnet mask, the 27 bits on the left are set to one. In this case, they talk about the length of the subfix prefix of 27 bits and indicate through a slash ( / sign) after the base address.

An example of writing IP address 172.16.0.1/12 using classless addressing:

The set of all addresses corresponds to a zero subnet mask and is denoted by / 0, and a specific IPv4 address corresponds to a subnet mask with a prefix length of 32 bits, denoted by / 32.

To simplify routing tables, you can combine address blocks by specifying one large block instead of a number of small ones. For example, 4 adjacent networks of class C (4 × 255 addresses, mask 255.255.255.0 or / 24) can be combined, from the point of view of routers far from them, into one / 22 network. Conversely, networks can be broken up into smaller subnets, and so on.

The standard is a mask in the form of a continuous sequence of ones and a continuous sequence of zeros. Only for such masks the resulting sets of IP addresses will be contiguous. However, inverse masks are also widespread (invers mask, wildcard mask), which are not required to contain consecutive ones or zeros. The reverse mask is used to form the ACL rules.

Possible masks [edit | edit wiki text]

1. IPv4 CIDR

The number of subnet addresses is not equal to the number of possible nodes. A zero IP address is reserved for identifying a subnet, the last one as a broadcasting address, so in real-life networks, the number of nodes is two less than the number of addresses.