Subnet & IP calculation

CIDR notation (like 192.168.1.0/24) states an IP address alongside a prefix length — the number of leading bits that identify the network, with the remaining bits identifying hosts within it. A /24 reserves 8 bits for hosts, giving 256 total addresses, of which 254 are usable in a conventional IPv4 subnet (the first address identifies the network itself, the last is the broadcast address). The subnet mask (255.255.255.0 for a /24) is just another way of writing the same prefix length as a 32-bit pattern of 1s and 0s.

VLSM — variable-length subnet masking — is what lets you split one address block into subnets of different sizes instead of forcing every subnet to be equal. A network with one segment needing 100 hosts and another needing 6 wastes a lot of address space if both get identical fixed-size blocks; VLSM lets you allocate a /25 to the first and a /29 to the second, matching actual need. This is standard practice in real network design, and it's also the calculation most people find hardest to do reliably by hand, since it means working out host counts backwards from a target prefix length repeatedly.

IPv6 changes some of the arithmetic: addresses are 128 bits instead of 32, prefix notation works the same way (a /64 is the typical subnet size for a LAN segment), but there's no broadcast address concept — IPv6 uses multicast for the equivalent functionality. The address space is large enough that IPv6 subnetting is rarely about scarcity the way IPv4 is; it's more about organizing hierarchy cleanly. Subnetic, SubnetMaskCalculator, and NetLoom each cover this ground for iOS/macOS, from single-address lookups to VLSM planning to batch CIDR aggregation across pasted address lists.