Understanding subnet sizes is one of those networking skills that feels small at first, but it quickly becomes essential when you are designing a home lab, configuring cloud infrastructure, securing a business network, or troubleshooting connectivity. A /28 subnet is a common choice when you need a small block of IP addresses without wasting too much address space.
TLDR: A /28 subnet contains 16 total IP addresses. In a traditional IPv4 network, 14 of those addresses are usable for hosts because one address is reserved for the network and one is reserved for broadcast. The subnet mask for a /28 is 255.255.255.240. A /28 is often used for small networks, point-of-presence deployments, firewall zones, and small public IP allocations.
What Does “/28” Mean?
The /28 notation comes from CIDR, which stands for Classless Inter-Domain Routing. CIDR notation tells you how many bits of an IP address are used for the network portion. Since an IPv4 address is 32 bits long, a /28 means that the first 28 bits identify the network, leaving the remaining 4 bits for host addresses.
Those 4 host bits are the key to calculating the size of the subnet. Each bit can be either 0 or 1, so the formula is:
2number of host bits = total addresses
For a /28 subnet:
24 = 16 total IP addresses
That is the quick mathematical reason a /28 subnet contains 16 IP addresses.
Total IP Addresses vs. Usable IP Addresses
In IPv4 networking, the total number of addresses in a subnet is not always the same as the number of addresses you can assign to devices. In most traditional IPv4 subnets, two addresses are reserved:
- Network address: Identifies the subnet itself.
- Broadcast address: Used to send traffic to all hosts on the subnet.
So, while a /28 has 16 total addresses, it usually has:
16 total addresses – 2 reserved addresses = 14 usable host addresses
This means you can typically assign addresses to 14 devices, interfaces, servers, routers, firewalls, or other network hosts within a /28 IPv4 subnet.
Example of a /28 Subnet
Let’s say you have the subnet:
192.168.1.0/28
This subnet includes the following IP address range:
- Network address: 192.168.1.0
- Usable host range: 192.168.1.1 through 192.168.1.14
- Broadcast address: 192.168.1.15
That gives you exactly 16 total addresses, with 14 usable addresses. The next /28 subnet would begin at 192.168.1.16/28, then the next at 192.168.1.32/28, and so on. This pattern happens because each /28 block increases by 16 addresses.
The Subnet Mask for a /28
A /28 subnet corresponds to the subnet mask:
255.255.255.240
This mask might look a bit mysterious if you are used to seeing common masks like 255.255.255.0. The final octet, 240, is what creates the 16-address block size. In binary, 240 is written as:
11110000
The four 1s represent the network bits in the last octet, and the four 0s represent the host bits. Since there are four host bits, the subnet can contain 16 possible addresses.
A useful shortcut is to subtract the last subnet mask octet from 256:
256 – 240 = 16
That result tells you the block size. So with a /28, subnets move in increments of 16.
Common /28 Subnet Ranges
If you are subnetting a larger network, such as 192.168.1.0/24, into /28 blocks, the ranges would look like this:
- 192.168.1.0/28: 192.168.1.0 to 192.168.1.15
- 192.168.1.16/28: 192.168.1.16 to 192.168.1.31
- 192.168.1.32/28: 192.168.1.32 to 192.168.1.47
- 192.168.1.48/28: 192.168.1.48 to 192.168.1.63
- 192.168.1.64/28: 192.168.1.64 to 192.168.1.79
Each subnet contains 16 total addresses. The first address in each block is the network address, and the last address is the broadcast address.
When Would You Use a /28 Subnet?
A /28 subnet is small, but that is often exactly what makes it useful. It provides enough addresses for a limited group of systems while keeping the network compact and manageable. In IPv4, where public address space is limited and valuable, using a /28 can be much more efficient than assigning a larger subnet unnecessarily.
Common uses for a /28 include:
- Small office networks: Enough addresses for a router, printer, access point, and several workstations.
- Public IP allocations: Internet service providers may assign /28 blocks to businesses that need multiple public IPs.
- Firewall DMZ networks: A /28 works well for exposing a small number of public-facing servers.
- Cloud networking: Small subnets can separate services, environments, or security zones.
- Lab environments: A /28 is ideal for practicing subnetting without consuming large ranges.
For example, a company might use a /28 public subnet for a firewall, VPN appliance, web server, mail gateway, monitoring server, and a few spare addresses. The subnet is large enough to be practical but small enough to avoid waste.
How a /28 Compares to Other Subnet Sizes
To understand a /28 better, it helps to compare it with nearby subnet sizes:
- /29: 8 total addresses, usually 6 usable
- /28: 16 total addresses, usually 14 usable
- /27: 32 total addresses, usually 30 usable
- /26: 64 total addresses, usually 62 usable
- /24: 256 total addresses, usually 254 usable
The smaller the CIDR number, the larger the subnet. So a /27 is larger than a /28, while a /29 is smaller. This can seem backward at first, but it makes sense when you remember that the slash number represents the number of network bits. More network bits leave fewer host bits.
What About IPv6?
The answer above applies mainly to IPv4. IPv6 uses 128-bit addresses and does not use broadcast addresses in the same way IPv4 does. Technically, an IPv6 /28 would contain an enormous number of addresses because there would be 100 host bits remaining. However, in practical IPv6 design, subnets such as /64 are far more common for local networks.
So if someone asks, “How many IP addresses are in a /28 subnet?” they are almost always referring to IPv4 unless IPv6 is specifically mentioned.
Quick Calculation Method
If you want a fast way to calculate subnet sizes, use this process:
- Start with 32, the number of bits in an IPv4 address.
- Subtract the CIDR prefix: 32 – 28 = 4.
- Use the result as the exponent: 24 = 16.
- Subtract 2 for typical usable IPv4 hosts: 16 – 2 = 14.
This method works for any IPv4 subnet. For example, a /27 leaves 5 host bits, so it has 25, or 32, total addresses. A /29 leaves 3 host bits, so it has 8 total addresses.
Final Answer
A /28 subnet has 16 total IP addresses. In a standard IPv4 network, 14 addresses are usable for hosts because the first address identifies the network and the last address is reserved for broadcast. Its subnet mask is 255.255.255.240, and its address blocks increase in steps of 16.
In short, a /28 is a compact and efficient subnet size. It is large enough for small network segments, public server groups, and firewall zones, but small enough to conserve IPv4 address space. If you remember just one thing, remember this: /28 equals 16 total addresses and usually 14 usable IPv4 addresses.
