The /16 subnet, explained.

The /16 subnet uses 255.255.0.0 as its subnet mask — meaning the first 16 bits of every address identify the network, and the remaining 16 bits identify the host within that network. That gives you 65,536 total addresses (65,534 usable on standard RFC math, after subtracting the network and broadcast addresses).

The wildcard mask — the bitwise inverse of the subnet mask — is 0.0.255.255. Wildcards are what Cisco access-control lists and OSPF area definitions use instead of subnet masks; the "1" bits mark "don't care" positions. For a /16, that leaves 16 don't-care host bits.

To find the network address for any IP in a /16 block, perform a bitwise AND between the IP and the subnet mask. To find the broadcast, OR the network address with the wildcard. Modern tools — like our subnet calculator — do this in microseconds, but the underlying mechanics are straightforward binary arithmetic.

A /16 is the classic large-network allocation. 65,536 addresses, often used as a corporate site aggregate or as the parent CIDR of an AWS VPC, Azure VNet, or GCP custom subnet. The maximum VPC size in AWS is /16.

Cloud-provider quirks matter at every prefix size: AWS and Azure reserve 5 IPs per subnet, GCP reserves 4, and OCI reserves 3. So a /16 on standard RFC math gives you 65,534 usable hosts, but on AWS or Azure that drops to 65,531. The capacity-planning gap bites hardest at small prefixes (a /28 has 14 usable on paper, only 11 on AWS) but exists at every size. Our cloud-aware calculator applies the right math automatically.