Every bit you move the subnet mask boundary trades in one direction: more subnets, fewer hosts per subnet, or the reverse. It's a fixed-size address pool being divided differently, not created or destroyed.
Network engineering designs how devices are addressed, connected, and routed across a network. Subnetting — dividing a larger address block into smaller subnets — is one of the most fundamental everyday tasks, and it involves a strict tradeoff: the same fixed pool of address bits can be allocated toward more subnets or more hosts per subnet, but never both simultaneously, since the total number of address bits never changes.
An IPv4 address has 32 bits total. The subnet mask (expressed as a prefix length, e.g., /24) divides those bits into a network portion and a host portion. Moving the boundary — increasing the prefix length — takes bits away from the host portion and gives them to the network portion, which is exactly why it simultaneously increases the number of possible subnets and decreases the number of usable host addresses per subnet, as shown in the visualization above.
Within any subnet, the all-zeros host address is reserved as the network address (identifying the subnet itself), and the all-ones host address is reserved as the broadcast address (reaching every device on that subnet at once) — neither can be assigned to an individual device, which is why usable host count is always 2^(host bits) − 2, not the full 2^(host bits).
Choosing subnet sizes correctly avoids two opposite failure modes: subnets too large waste address space and put too many devices on one broadcast domain (increasing broadcast traffic and reducing performance); subnets too small run out of usable addresses as a location grows, forcing a disruptive re-addressing later. Matching subnet size to actual expected host count, with reasonable growth room, is a core practical network design skill.
A /30 leaves 2 host bits, giving 4 total addresses in that block — but 2 are always reserved (network address and broadcast address), leaving exactly 2 usable addresses. This makes /30 subnets a very common choice for point-to-point links between two routers, where exactly 2 addresses are needed and no more.
No — subnetting only reorganizes how a fixed-size address block is divided; it never increases or decreases the total number of addresses in that block. It's purely a reallocation between 'more subnets' and 'more hosts per subnet.'
Once a subnet's usable host addresses are exhausted, no new devices can be added without either creating an additional subnet for the overflow or re-addressing the whole subnet to a larger block — both disruptive changes, which is why initial subnet sizing should account for reasonable future growth, not just current device count.
Try our STEM Learning Studio
More calculators, simulators, and guides for this discipline.