Subnetting: From First Principles to VLSM

Subnetting is one of those skills that seems intimidating until it clicks, and then it feels obvious. This post builds from first principles — binary representation, masks, and block sizes — through variable-length subnet masking (VLSM) and summarization.

IP Address Structure

An IPv4 address is 32 bits divided into two logical parts: network and host.

192.168.10.25 = 11000000.10101000.00001010.00011001

The subnet mask defines the split. A 255.255.255.0 mask (or /24 in CIDR) means the first 24 bits identify the network, the last 8 identify the host.

Address:  192.168.10.25   11000000.10101000.00001010.00011001
Mask:     255.255.255.0   11111111.11111111.11111111.00000000
                          ^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^
                          Network portion           Host portion

CIDR Prefix Lengths

The prefix length (/x) is just the count of consecutive 1-bits in the mask.

Usable hosts = 2^(32−prefix) − 2, subtracting network and broadcast addresses.
Exception: /31 is used for point-to-point links (RFC 3021) — both addresses are usable.

The Block Size Trick

Block size = 256 − last octet of mask.

For a /26 (mask 255.255.255.192):

Block size = 256 - 192 = 64

Subnets in the last octet increment by 64:

192.168.1.0/26    → hosts .1–.62,  broadcast .63
192.168.1.64/26   → hosts .65–.126, broadcast .127
192.168.1.128/26  → hosts .129–.190, broadcast .191
192.168.1.192/26  → hosts .193–.254, broadcast .255

To find which subnet an address belongs to: divide the host octet by the block size, take the floor, multiply back.

Which subnet is 192.168.1.100/26?
100 / 64 = 1 (floor)
1 × 64   = 64  → subnet is 192.168.1.64/26

Calculating Network, Broadcast, and Host Range

Given 10.0.5.87/28:

1. Block size: 256 − 240 = 16
2. Network address: 87 / 16 = 5 (floor) → 5 × 16 = 80 → 10.0.5.80
3. Broadcast: network + block − 1 → 80 + 16 − 1 = 95 → 10.0.5.95
4. Host range: 10.0.5.81 – 10.0.5.94

VLSM: Allocating Subnets Efficiently

Variable-Length Subnet Masking lets you carve a block into subnets of different sizes. The key rule: allocate largest subnets first to avoid fragmentation.

Example

Assign subnets from 172.16.0.0/24 for:

Step 1 — LAN A (100 hosts): needs /25 (126 usable)

172.16.0.0/25   hosts: .1–.126   broadcast: .127

Step 2 — LAN B (50 hosts): needs /26 (62 usable)

172.16.0.128/26  hosts: .129–.190  broadcast: .191

Step 3 — LAN C (25 hosts): needs /27 (30 usable)

172.16.0.192/27  hosts: .193–.222  broadcast: .223

Step 4 — WAN links (2 hosts each): /30 (2 usable)

172.16.0.224/30  hosts: .225–.226  broadcast: .227
172.16.0.228/30  hosts: .229–.230  broadcast: .231

Remaining space: 172.16.0.232/29 through 172.16.0.255 — available for future growth.

Summarization

Route summarization (aggregation) collapses multiple specific prefixes into one covering advertisement. This is what BGP route aggregation and OSPF inter-area summarization both rely on.

To summarize a set of prefixes:

1. Write all network addresses in binary.
2. Find the longest common bit prefix.
3. The summary prefix length equals that common bit count.

Example

Summarize 10.1.4.0/24, 10.1.5.0/24, 10.1.6.0/24, 10.1.7.0/24:

10.1.4.0  = 00001010.00000001.00000100.00000000
10.1.5.0  = 00001010.00000001.00000101.00000000
10.1.6.0  = 00001010.00000001.00000110.00000000
10.1.7.0  = 00001010.00000001.00000111.00000000
                                      ^^
                               first difference at bit 23

Common prefix: 22 bits → summary is 10.1.4.0/22.

Watch out for over-summarization. 10.1.4.0/22 also covers 10.1.4.0–10.1.7.255. If you only own .4.0–.7.0/24 but advertise /22, you’ll black-hole traffic for any addresses in that range you don’t actually have routes for.

Private Address Space (RFC 1918)

And 100.64.0.0/10 (RFC 6598) is reserved for carrier-grade NAT — you’ll see it on ISP CPE and in some cloud provider transit ranges.

Quick Mental Math Tips

• Powers of 2: memorize 2^1 through 2^8 (2, 4, 8, 16, 32, 64, 128, 256). Everything else follows.
• Host count shortcut: for a /x prefix in the last octet, hosts = block size − 2.
• Subnet count: dividing a /24 into /26 gives 2^(26−24) = 4 subnets.
• /31 for P2P links: saves address space on transit links; supported by all modern platforms. Use it.
• /32 for loopbacks: always. Never put a loopback on a /24 — it wastes 254 addresses and can cause unnecessary routing entries.

Related Topics

Subnetting is the foundation for:

• OSPF area design — summarization at ABRs relies on contiguous address blocks
• BGP prefix filtering — prefix-lists match on prefix + le/ge operators
• ACL design — wildcard masks are the inverse of subnet masks (255 − mask octet)
• IPv6 — the same CIDR logic applies, just with 128-bit addresses and /64 as the standard LAN prefix