/16 vs /17 — Subnet Comparison

A /16 subnet is larger than a /17. Every additional bit in the prefix halves the address space — the 1-bit difference between these two means /16 has 21 = 2 times as many addresses.

/16

65K IPs — standard VPC or site block

Full reference →
Total IPs 65,536
Usable Hosts 65,534
Subnet Mask 255.255.0.0
Wildcard Mask 0.0.255.255

Typical Uses

  • AWS / Azure VPC or VNet CIDR
  • Enterprise campus or data centre segment
  • Docker overlay network
/17

32K IPs — half a /16

Full reference →
Total IPs 32,768
Usable Hosts 32,766
Subnet Mask 255.255.128.0
Wildcard Mask 0.0.127.255

Typical Uses

  • Half of a /16 VPC split public/private
  • Large campus segment

Key Differences

more IPs in /16 than /17
2
/17 subnets fit inside one /16
1
bit of difference in prefix length

How 2 /17 Subnets Divide a /16

Example using 10.0.0.0/16 as the parent block.

# CIDR Network First Usable Last Usable Broadcast Hosts
1 10.0.0.0/17 10.0.0.0 10.0.0.1 10.0.127.254 10.0.127.255 32,766
2 10.0.128.0/17 10.0.128.0 10.0.128.1 10.0.255.254 10.0.255.255 32,766

FAQ

What is the difference between /16 and /17?

A /16 has 65,534 usable hosts and a /17 has 32,766. The subnet masks differ: /16 uses 255.255.0.0 while /17 uses 255.255.128.0. Every additional bit in the prefix halves the number of addresses — so the 1-bit gap means /16 is exactly 2× larger.

How many /17 subnets fit in a /16?

Exactly 2 /17 subnets fit perfectly inside one /16 with no wasted space. To split a /16 into /17s, just increment the last 1 bit of the network address for each new subnet.

Which should I choose?

/16 is typically used for: Cloud VPC CIDR, campus network. /17 is better for: Large segment within a /16. Choose the smallest prefix that comfortably fits your host count — over-allocating wastes address space, but under-allocating means painful renumbering later.