VLSM preventing IP address loss

Concept

Variable Length Subnet Mask (VLSM) lets one network address be divided into subnets of different sizes by using a different mask on each subnet. In fixed-length subnetting (FLSM) every subnet has the same size, so small subnets (like a router-to-router link needing 2 addresses) waste a whole large block. VLSM gives each subnet only the block it needs, so unused addresses stay available.

Example (assumed)

Organisation has 192.168.10.0/24 (256 addresses). Requirements:

• LAN A: 100 hosts
• LAN B: 50 hosts
• LAN C: 25 hosts
• WAN link 1 (R1-R2): 2 hosts
• WAN link 2 (R2-R3): 2 hosts

With FLSM: the largest LAN needs 100 hosts, so every subnet must be /25 (126 hosts). Five subnets of 128 addresses need 640 addresses, which is more than the /24 holds. Each WAN link wastes 126 of 128 addresses.

With VLSM (allocate largest first):

Working: 2^7 - 2 = 126 >= 100 (so 7 host bits, /25); 2^6 - 2 = 62 >= 50 (/26); 2^5 - 2 = 30 >= 25 (/27); 2^2 - 2 = 2 (/30).

        LAN A             LAN B            LAN C
   192.168.10.0/25   .128/26          .192/27
        |                |                |
      [R1]---.224/30---[R2]---.228/30---[R3]

How VLSM prevents loss of addresses

• Used addresses: 128 + 64 + 32 + 4 + 4 = 232 out of 256; the remaining 24 (192.168.10.232 to .255) are free for future subnets.
• With FLSM the WAN links alone would waste 2 x 126 = 252 addresses; with VLSM each /30 wastes only 2 addresses (network and broadcast).
• Addresses saved can be reused, which delays the need for another network block.
• VLSM also supports route summarisation and hierarchical design, but needs a classless routing protocol (RIPv2, OSPF, EIGRP, BGP) because the mask is carried in routing updates.