subnetzmaske

Subnet Mask

In short: A bit mask (e.g. 255.255.255.0) that determines, for an IP address, which part identifies the network and which part the individual device (host).

In more detail: Often also given in CIDR notation (/24 instead of 255.255.255.0), where the number states how many bits from the left are reserved for the network. This directly determines how many hosts fit into a subnet.

In Depth

A subnet mask works like a stencil laid bit by bit over the IP address: every bit set to 1 in the mask belongs to the network part, every bit set to 0 to the host part. With 255.255.255.0 (in binary: 24 ones followed by 8 zeros), the first three octets are fixed (the network) and the last octet is free for individual devices — this gives 2^8 = 256 possible values, i.e. 254 usable host addresses after subtracting the network and broadcast addresses.

CIDR notation (/24) states the number of network bits directly and is the more common notation today, because it’s more compact and allows more flexible divisions not tied to byte boundaries (e.g. /27 for only 30 usable hosts, sensible for a very small subnet). Rule of thumb: the larger the number after the slash, the SMALLER the subnet (more bits for the network = fewer bits left for hosts) — /30, for example, allows only 2 usable host addresses and is often used for pure point-to-point links between two routers.

See also: Subnet, IP addresses