In all of the previous examples of subnetting, notice that the same subnet mask was applied for all the subnets. This means that each subnet has the same number of available host addresses.
As illustrated in Figure 1, traditional subnetting creates subnets of equal size. Each subnet in a traditional scheme uses the same subnet mask. As shown in Figure 2, VLSM allows a network space to be divided in unequal parts. With VLSM the subnet mask will vary depending on how many bits have been borrowed for a particular subnet, thus the “variable” part of the VLSM.
VLSM subnetting is similar to traditional subnetting in that bits are borrowed to create subnets. The formulas to calculate the number of hosts per subnet and the number of subnets created still apply. The difference is that subnetting is not a single pass activity. With VLSM, the network is first subnetted, and then the subnets are subnetted again. This process can be repeated multiple times to create subnets of various sizes.