You just finished setting up your office network. Worth adding: twenty-three computers, a printer, a couple of servers, and a few IP phones. You figure a /28 would cover it — after all, that's 16 addresses, and you're only using 27 devices, right?
Except a /28 gives you 16 total* addresses, which means only 14 usable ones for your devices. You're already short Simple, but easy to overlook..
This is the kind of mistake that gets made more often than you'd think. Subnet sizing trips up a lot of people, and the math isn't intuitive when you first encounter it. A /27 sits in that awkward middle ground — big enough to handle a decent number of devices, small enough that you might assume it's just a /28 with a few extra addresses tacked on Which is the point..
It's not. And understanding exactly how many addresses a /27 contains, why those numbers matter, and how to work with them properly is the kind of knowledge that saves you from a scramble later That alone is useful..
What Is a /27 Subnet?
Let's start with CIDR notation — that's the "/27" part. CIDR, or Classless Inter-Domain Routing, is how we express IP address blocks and their corresponding subnet masks without relying on the old classful system (Class A, B, C networks, if you remember those from older networking history).
Worth pausing on this one And that's really what it comes down to..
When you see "/27," here's what it means: 27 bits are reserved for the network portion of the address. The remaining 5 bits are available for host addresses within that network.
IP addresses are 32 bits total, split into four octets of 8 bits each. So a /27 tells you how many of those 32 bits define the network, and how many are left over for individual devices.
The math breaks down like this: 2 to the power of the remaining host bits gives you the total number of addresses in the block. Since 5 bits are left for hosts, you get 2^5 = 32 total addresses per /27 network Worth keeping that in mind..
How the Subnet Mask Translates
The /27 subnet mask translates to 255.224 in dotted decimal notation. 255.In practice, 255. If you've ever wondered why the last octet jumps in specific increments — 0, 32, 64, 96, 128, and so on — it's because the "224" in that last octet represents which bits are "borrowed" from the host portion.
Each /27 block spans 32 addresses in the final octet. So your networks would be:
- 192.168.Even so, 1. Now, 0/27 (addresses . On the flip side, 0 through . Because of that, 31)
-
- 168.But 1. 32/27 (addresses .32 through .Because of that, 63)
-
- 168.1.64/27 (addresses .64 through .
And so on, incrementing by 32 each time Surprisingly effective..
Why It Matters
Here's the practical reality: choosing the wrong subnet size creates real problems.
Go too small, and you run out of IP addresses for your devices. You end up renumbering your network — which sounds simple until you realize every device with a static IP needs to be updated, every DHCP reservation, every firewall rule referencing those addresses Simple, but easy to overlook. And it works..
Go too large, and you're wasting address space. On top of that, in enterprise environments or ISP allocations, this matters. Even in smaller setups, it's a habit worth getting right.
A /27 gives you 30 usable addresses (more on the "usable" part in a moment). That makes it a solid choice for small office networks, server rooms with a handful of machines, or network segments where you want some breathing room without committing a /24 (256 addresses) unnecessarily But it adds up..
Not obvious, but once you see it — you'll see it everywhere.
Understanding these numbers also matters when you're reading network diagrams, troubleshooting connectivity issues, or having conversations with IT colleagues. If someone says "we need a /27 for that segment," you should immediately know what that implies for your address planning.
How It Works: The Breakdown
Total Addresses vs. Usable Addresses
Here's the part that catches people: a /27 has 32 total addresses, but only 30 are usable for devices.
Why the difference? Two addresses in every IP block are reserved and can't be assigned to hosts:
-
The network address — the first address in the block, where all host bits are set to 0. This identifies the network itself, not any specific device.
-
The broadcast address — the last address in the block, where all host bits are set to 1. This is used to send a packet to every device in the network simultaneously Practical, not theoretical..
So when you're counting up the devices you need to support, you need to account for these two reserved addresses. A /27 with 30 usable addresses means you can assign IPs to 30 devices, not 32.
Quick Reference Table
Here's how /27 compares to nearby subnet sizes:
| CIDR | Total Addresses | Usable Hosts | Subnet Mask |
|---|---|---|---|
| /28 | 16 | 14 | 255.Practically speaking, 255. 255.Because of that, 240 |
| /27 | 32 | 30 | 255. 255.That's why 255. 224 |
| /26 | 64 | 62 | 255.255.255. |
The pattern is consistent: each step up (lower number) doubles the address space. Each step down (higher number) halves it.
Working Through an Example
Let's say you're assigning a /27 to a small office network. 113.Practically speaking, 0. Your ISP has given you the range 203.0/27 The details matter here. Which is the point..
Your usable range would be 203.0.113.1 through 203.0.113.30 The details matter here..
- Network address: 203.0.113.0 (reserved)
- First usable IP: 203.0.113.1 (often assigned to a gateway or router)
- Last usable IP: 203.0.113.30
- Broadcast address: 203.0.113.31 (reserved)
You'd typically assign one of the usable addresses (usually the first or last) to your default gateway, leaving 29 addresses for workstations, printers, and other devices.
Common Mistakes and What People Get Wrong
**Thinking a /27 has 32
usable addresses.** To revisit, two addresses are always reserved. If you plan for 32 devices on a /27, you'll find yourself one or two IPs short when you try to assign them Less friction, more output..
Confusing total addresses with usable addresses when reading documentation. Some documentation lists the raw block size without subtracting the network and broadcast addresses. Always clarify which number you're working with before making plans.
Misaligning subnet boundaries. A /27 must start on a multiple of 32 in the final octet (0, 32, 64, 96, 128, 160, 192, 224). If you try to use a range like 203.0.113.10 to 203.0.113.41 as a /27, it won't work properly because the boundaries don't align. This is a common error when manually planning subnets without using a calculator.
Forgetting about future growth. A /27 might seem like plenty today, but if you're adding devices, virtual machines, or IoT equipment, that headroom can disappear quickly. It's worth considering whether you might need a /26 instead, even if it feels like overkill initially And it works..
Where You'll See /27s in Practice
Point-to-point links. Network engineers sometimes use /27s (or even /30s and /31s) for connections between routers because only a few addresses are needed. While /30 is more common for this purpose, /27s appear in scenarios where some flexibility is desired Worth keeping that in mind..
VPN tunnels and remote networks. Branch offices connecting back to a headquarters via VPN often get /27 allocations. It's enough for staff workstations, a printer or two, and perhaps a phone system, without wasting larger blocks that could be used elsewhere.
Small business networks. A typical small business with 10–25 employees fits comfortably within a /27. There's room for growth, and the block is easy to manage without complex subnetting Took long enough..
DMZ segments. When creating a demilitarized zone for publicly accessible servers, a /27 provides enough room for web servers, mail servers, and related infrastructure while keeping the segment isolated from the internal network Still holds up..
Calculating It Yourself
If you want to verify or calculate /27 ranges without relying on a table, here's the formula:
- Number of addresses: 2^(32 − prefix length)
- For /27: 2^(32−27) = 2^5 = 32 total addresses
- Usable hosts: Total − 2 = 30
To find the network address of a given IP, you can use bitwise AND with the subnet mask. Think about it: 113. And 255. 0.For 203.41 with a /27 mask (255.255.
- 203.0.113.41 in binary (last octet): 00101001
- 255.255.255.224 in binary (last octet): 11100000
- AND result: 00100000 = 32
So 203.Here's the thing — 113. 113.Because of that, 113. 0.And 0. In practice, 32/27 network, with usable addresses from 203. Worth adding: 41 falls within the 203. 0.Which means 113. Think about it: 33 to 203. And 0. 62 Most people skip this — try not to..
Once you've practiced this a few times, the math becomes intuitive.
Tools That Make It Easier
While understanding the math is valuable, you don't need to do it by hand every time. Several tools can speed up subnet calculations:
- Online subnet calculators — Websites like subnet-calculator.com or others let you input a CIDR block and instantly see the network range, usable hosts, and broadcast address.
- Command-line tools — On Linux or macOS,
ipcalcprovides quick calculations. On Windows, PowerShell has built-in networking commands that can help. - Network management software — Enterprise tools often include subnet planning features that visualize your address space and flag overlaps or inefficient allocations.
These tools are especially helpful when you're managing multiple subnets and need to avoid conflicts or gaps in your addressing scheme Simple, but easy to overlook..
The Bigger Picture
Understanding /27s is one piece of a larger puzzle. Once you're comfortable with subnetting basics, you can move on to more advanced topics like:
- Variable Length Subnet Masking (VLSM) — Using different subnet sizes within the same network to make efficient use of your address space.
- Supernetting and CIDR aggregation — Combining multiple smaller networks into larger blocks for simpler routing.
- IPv6 subnetting — The concepts carry over, but the address space is vastly larger.
Each of these builds on the same foundation: understanding how prefixes determine the size of a network and how addresses are allocated within it.
Wrapping Up
A /27 subnet gives you 32 total addresses and 30 usable hosts, sitting comfortably between a /28 (14 usable) and a /26 (62 usable). It's a practical size for small networks, branch offices, VPN endpoints, and isolated segments where you want enough room without overcommitting address space.
The key things to remember are that two addresses are always reserved (the network and broadcast addresses), subnet boundaries must align on multiples of 32 in the final octet, and understanding these details makes you more effective whether you're planning a network, reading documentation, or troubleshooting an issue.
Subnetting isn't glamorous, but it's one of those foundational skills that pays off every time you work with networks. The /27 is a perfect example of a concept that's simple once you understand it but confusing if you don't. Take the time to internalize the pattern, and the rest of subnetting will follow naturally Simple as that..