Visual Subnet Calculator

How a Visual Subnet Calculator Turns CIDR Math Into a Map

Every IP network is just a range of numbers, but the tool represents that range as a single block you can see and click. Enter a starting address and a prefix length, and it draws the whole block as a rectangle you can work with directly. Click Divide and the rectangle splits into two equal halves, each with its own address details; click Join to merge two adjacent pieces back together. The whole point is to split and join subnets visually instead of tracking every change in a spreadsheet.

Reading Network Address, Netmask, and Range of Addresses

Each block exposes a handful of fields you'd otherwise work out by hand: the address that starts it, the mask bits that define its size (shown as something like /26), and the range of addresses it spans. Between the first and last address sit the useable IPs — the ones you can actually assign to hosts — while usable hosts is simply how many of those exist.

The Subnet Mask Formula Behind Every Split

Behind the visualization, the math is straightforward binary arithmetic. Given a prefix length \(p\), the number of addresses in a block is:

$$ \text{Total Addresses} = 2^{(32 - p)} $$

Because the first address in a standard block is reserved as the network address and the last as the broadcast address, the usable count drops by two:

$$ \text{Usable Hosts} = 2^{(32 - p)} - 2 $$

That formula is what runs automatically every time you click Divide, instead of leaving you to work it out on paper.

Cloud Subnet Reserved Addresses: What an IP Calculator Needs to Know

Providers don't all reserve the same number of addresses per block, which is why a general-purpose calculator falls short once you're planning multi-provider infrastructure. Each cloud subnet carves out its own reserved addresses before you get a single usable host — AWS mode, Azure mode, and OCI mode each apply different rules.

AWS VPC Reservations

In AWS, the smallest block you can create is a /28, and five addresses are reserved out of every one: the block's starting address, an internal router, an internal DNS entry, one held for future use, and the final address. That's two more reservations than a standard subnet, so a /28 here yields 11 usable addresses rather than 14.

Azure Subnet Reservations

Azure reserves its addresses a little differently, and the platform won't let you go smaller than a /29. Several Azure services also require a dedicated block of their own rather than sharing space with your workloads:

A dns resolver endpoint, a firewall subnet, and a gateway subnet each get their own reserved block at the recommended size Azure publishes for that workload — undersizing any of them means re-carving your address plan later.

OCI Mode and Google VPC Considerations

This platform reserves only three addresses per subnet and allows blocks as small as a /30 — the tightest reservation model among the major providers. If you're planning across a multi-provider footprint, remember Google's platform uses its own reservation rules too, so matching whichever provider's mode the tool supports keeps your host counts accurate instead of optimistic.

Self-Hosting an IPAM Tool with Docker

Because this kind of tool does its math entirely client-side, most implementations can be self-hosted with a single container rather than relying on a public website for internal network planning. Pulling a prebuilt image and running it locally keeps your layout private, which matters if it maps to infrastructure you'd rather not expose.

Running the Container Locally

A typical setup exposes it on a local port with one command:

docker run -d -p 8080:8080 visualsubnetcalc

From there, opening the address in a browser gives you the same split, merge, and visualization tools you'd get from a hosted version — just running entirely on infrastructure you control.