IPv4 and IPv6 Subnet Calculator

A prefix length such as /24 says how many leading bits of an address name the network. The rest name a host inside it. That notation is CIDR, set out in RFC 4632, and it replaced the old class A, B and C boundaries.

The network address is the first address in the block and names the subnet itself. The broadcast address is the last one and reaches every host on the subnet. Neither is given to a device, which is why a /24 holds 256 addresses and 254 usable hosts.

The subnet mask is the prefix written as four numbers. The wildcard mask is the same mask inverted, and it is what Cisco access lists and OSPF network statements ask for. Typing the wrong one of the two into a firewall rule is a common way to open far more than was intended.

The address type line says whether the address is public, private (RFC 1918), shared carrier space (RFC 6598), loopback, link-local or reserved for documentation. It is worth a look before a range goes into a public DNS record or an allow list.

Put a longer prefix in the second box to divide the network. Every extra bit doubles the number of subnets and halves the size of each: a /24 becomes two /25, four /26 or eight /27. The table lists each subnet with its first host, last host and broadcast address, ready to copy into an address plan.

Size subnets for the hosts they will hold a few years from now, and keep servers, users, guests and management interfaces in separate ones. A subnet boundary is where a firewall rule can be placed. A flat network gives an intruder every machine once they have one.

IPv6 has no broadcast address and reserves no host addresses the way IPv4 does. A single LAN is a /64 by convention, because the interface identifier is 64 bits (RFC 4291), so planning is about how many /64 networks a site prefix holds and not about hosts per subnet. For an IPv6 prefix shorter than /64 the result shows that count.