Subnet Calculator (IP / CIDR)

An IPv4 address is 32 bits, written as four numbers from 0 to 255. The mask says how many of those bits, counted from the left, name the network; the rest number the hosts (devices) inside it. Writing /24 after an address (CIDR notation) means the first 24 bits are the network.

The formula

network address = IP AND mask
broadcast address = network OR wildcard
wildcard = 255.255.255.255 − mask
addresses = 2^(32 − prefix)
usable hosts = 2^(32 − prefix) − 2 (a /31 has 2, a /32 has 1)
first host = network + 1, last host = broadcast − 1
subnets from borrowing b bits = 2^b

Worked example

Take 192.168.1.10/24. In binary the address is 11000000.10101000.00000001.00001010 and the /24 mask is 24 ones followed by 8 zeros, which is 255.255.255.0. Keeping the first 24 bits and zeroing the last 8 gives the network 192.168.1.0. Setting the last 8 bits to 1 gives the broadcast 192.168.1.255. There are 2^8 = 256 addresses, and 256 − 2 = 254 usable hosts, from 192.168.1.1 to 192.168.1.254. The wildcard mask is 255.255.255.255 − 255.255.255.0 = 0.0.0.255. The first octet, 192, makes it a historic class C address, and it sits in 192.168.0.0/16, a private range (RFC 1918).

Splitting a network

To cut 192.168.1.0/24 into 4 subnets, borrow 2 host bits, because 2^2 = 4. The new prefix is /24 + 2 = /26, each subnet has 2^(32 − 26) = 64 addresses and 62 usable hosts, and they start at .0, .64, .128 and .192. If you need a number that isn't a power of two, round up: 5 subnets need 3 bits, which makes 8.

By hosts it works the other way: 50 hosts plus the network and broadcast addresses need 6 host bits (2^6 = 64), so each subnet is a /26 and a /24 holds 2^(26 − 24) = 4 of them.

VLSM: subnets of different sizes

VLSM sizes each subnet for its own hosts. Sort the needs from biggest to smallest, give each the smallest block that fits, and place them one after another, each starting on a multiple of its own size. In 192.168.1.0/24, needs of 50, 20, 10 and 2 hosts become 192.168.1.0/26, 192.168.1.64/27, 192.168.1.96/28 and 192.168.1.112/30. That uses 116 addresses and leaves 140 free as 192.168.1.116/30, 192.168.1.120/29 and 192.168.1.128/25.

Summarising (supernetting)

A summary route is the smallest single block that covers several networks. Write the lowest and highest addresses in binary and count how many leading bits they share. 192.168.0.0/24 to 192.168.3.0/24 share the first 22 bits, so they summarise to 192.168.0.0/22 with no extra addresses. If there are gaps, the summary also covers addresses you don't own, so the calculator also lists the exact blocks that cover only your networks.

Special IPv4 ranges

Every address outside these blocks is a public internet address.

Classes (historic)

Before CIDR arrived in 1993 (RFC 1519), addresses were split into classes by the first octet: A is 0–127 (default /8), B is 128–191 (/16), C is 192–223 (/24), D is 224–239 (multicast) and E is 240–255 (reserved). Routing no longer uses classes, but they still come up in exams and older kit, so the calculator shows the class for reference.

Subnet mask table

IPv6 in short

An IPv6 address is 128 bits, written as 8 groups of 4 hex digits. The short form (RFC 5952) drops leading zeros in each group and replaces the longest run of all-zero groups with ::, once. So 2001:0db8:85a3:0000:0000:8a2e:0370:7334 becomes 2001:db8:85a3::8a2e:370:7334. A normal IPv6 network is a /64; sites often get a /48 or /56, which hold 2^(64 − 48) = 65,536 or 2^(64 − 56) = 256 of them.

Ranges follow the IANA special-purpose address registries (RFC 6890) and the RFCs listed. The calculator works entirely in your browser; nothing you type is sent anywhere.