Internet Protocol version 4 (IPv4) Subnet Split Calculator

Split Summary

{{ subnetSummary.networkAddress }}/{{ cidr }} -> /{{ subnetSummary.childPrefix }}

{{ subnetSummary.totalNewNetworks.toLocaleString() }} Child Subnets {{ subnetSummary.childUsableCount.toLocaleString() }} Usable per Child {{ subnetSummary.totalReservedAddresses.toLocaleString() }} Reserved Total {{ subnetSummary.scopeLabel }} {{ hostFitBadge.label }}

IP address:

Enter any IPv4 host in the parent block, e.g. 192.168.10.45.

Current CIDR:

Select the existing parent prefix before you carve equal child subnets.

Planning mode:

Use child CIDR for mask-first planning; use count for powers-of-two targets.

Target CIDR:

Choose a deeper prefix through /30; each option shows the child count.

Child subnet count:

Only powers of two that fit the selected parent block are available.

{{ error }}

Needed hosts per child:

Whole number of usable hosts required per child subnet; use 0 for none.

hosts

Gateway host offset:

1 = first usable host, 2 = second; capped automatically for small child subnets.

usable slot

Preview rows:

Show the first 32-512 child subnets in the on-page ledger.

CSV header row:

Turn off only when appending rows into a file that already has headers.

CSV filename:

Use a short filename such as branch-a-subnet-ledger.csv.

Section	Field	Value	Detail	Copy
{{ row.section }}	{{ row.label }}	{{ row.value }}	{{ row.detail }}

#	Network	CIDR	Usable Range	Gateway	Broadcast	Usable	Copy
{{ row.index }}	{{ row.networkAddress }}	{{ row.cidr }}	{{ row.usableRange }}	{{ row.gatewayAddress }}	{{ row.broadcastAddress }}	{{ row.usableCount.toLocaleString() }}

Detailed child rows are unavailable for this split size. Use the planner table and forecast chart to choose a shallower split or narrower parent block.

Blue bars show usable hosts per child. The green line shows how many child subnets each deeper prefix creates. The selected target is highlighted, and the red guide marks your needed hosts per child when set.

Export to PDF Fullscreen

Include query parameters

Embed:

Customize embed code

Include query parameters

Wrap embed in collapsible toggle

Collapse panel by default

Hide card frame (bare iframe)

Loading behavior

Width

Height

Aspect ratio (width : height)

Max height (optional)

Collapsible heading

Collapsible description (optional)

Allow fullscreen

Referrer policy

Sandbox tokens

IPv4 subnet splitting takes one parent block and borrows host bits so that the same address space becomes several smaller, equal-size networks. That is the planning step behind carving one office range into VLANs, dividing a lab block across teams, or turning a larger site allocation into repeatable segments with clear boundaries.

This calculator starts from a practical input instead of a perfect worksheet. You can type any IPv4 address inside the parent block, choose the current prefix, and then plan the subdivision either by picking the child CIDR or by choosing how many equal child subnets you want. The page normalizes the entered address back to the parent network before it does the split.

A deeper prefix creates more child subnets, but each child repeats the same reserved edges. The page makes that tradeoff visible before you copy ranges into DHCP, router, or firewall work.

The output is meant for planning and handoff, not just arithmetic. You get a parent-block summary, a child-layout summary, a capacity verdict against an optional host target, a child-by-child ledger with gateway suggestions, a forecast chart, and JSON output that mirrors the same state.

The calculation stays in the browser, and the scope stays narrow on purpose. Every child subnet must be the same size, and the visible split choices stop at /30. That makes the page useful for equal subdivision and host-capacity checks, but not for mixed-size VLSM design or special /31 point-to-point treatment.

Technical Details

Classless Inter-Domain Routing, usually shortened to CIDR, treats an IPv4 network as a 32-bit value plus a prefix length. The current prefix defines the parent mask. This page converts the entered dotted-decimal address into a 32-bit integer, applies the parent mask to recover the canonical network address, and then uses the inverse of that mask to recover the parent broadcast boundary. That is why a host like 192.168.20.41 still resolves cleanly to a parent such as 192.168.20.0/24.

The split itself is straight prefix arithmetic. If the child prefix is deeper than the parent by d bits, the page creates 2^d equal child subnets. Each child contains 2^(32 - childPrefix) total addresses. In the visible workflow, usable host space is calculated with the conventional IPv4 rule of subtracting two addresses per child subnet for the network and broadcast edges. The UI therefore stops at /30 instead of trying to reinterpret /31 links under RFC 3021.

The planner table is built from the whole prefix ladder, not only the chosen split. For every deeper child prefix through /30, the page precomputes child count, block size, usable hosts per child, total reserved addresses, and an optional spare-or-short verdict against Needed hosts per child. The deepest row that still meets the host target becomes the page's best-fit recommendation.

The result surfaces also carry operational rules. Gateway host offset is clamped to the usable span of the chosen child subnet, Preview rows limits only the visible ledger, and the detailed ledger is suppressed when a split would create more than 4,096 child subnets. The range badge is a caution layer as well. It classifies the normalized parent against private, shared, link-local, documentation, benchmarking, multicast, reserved, and other special-use blocks, but that label is not a guarantee that a range is deployable in your environment.

Core Arithmetic

\begin{array}{lcl} N & = & A & M \\ B & = & N | ~ M \\ K & = & 2^{L child - L parent} \\ S & = & 2^{32 - L child} \\ U & = & S - 2 \\ G & = & U - H \end{array}

Meaning of IPv4 subnet split symbols and outputs
Symbol	Meaning here	Where you see it
`A`	Entered IPv4 address as a 32-bit value	Normalization of the parent block
`M`	Mask implied by the selected parent prefix	Parent network and broadcast recovery
`N`	Normalized parent network address	`Parent Block` and summary headline
`B`	Parent broadcast address	Boundary checks and parent usable span
`K`	Equal child subnet count produced by the deeper prefix	`Split Planner` rows and summary badges
`S`	Total addresses in each child subnet	`IPs / Child` and ledger rows
`U`	Usable host addresses per child subnet in this page	`Usable hosts` and chart bars
`G`	Host-fit gap against the requested host count	`Capacity Verdict` and `Host Fit`
`H`	Requested hosts per child	`Needed hosts per child`

Operational rules enforced by the IPv4 subnet split page
Behavior	What the page does	Why it matters
Equal-size model	Every child row uses the same child prefix and block size.	A good fit for repeated VLAN or segment plans, but not for VLSM.
`/30` cap	The child prefix ladder ends at `/30`.	The page stays aligned with the conventional network-and-broadcast host model.
Host-fit guidance	The page computes spare or short capacity for every deeper prefix when a host target is set.	You can see the deepest fitting equal split before copying any ranges.
Gateway clamp	The suggested gateway slot is reduced automatically if it exceeds usable hosts in the chosen child subnet.	Gateway suggestions stay inside the child range instead of pointing past the broadcast edge.
Ledger threshold	The detailed child ledger is withheld above 4,096 child subnets.	Large splits remain usable in summary and chart form without freezing the page.
Export scope	Preview limits shorten the on-screen ledger only. CSV, DOCX, and JSON still use the full rendered result set.	You can keep the page readable without losing the complete exported plan.

Everyday Use & Decision Guide

Pick the planning mode that matches the question you actually have. If you already know the child size you want, choose Choose child CIDR and work directly with the target prefix. If you know the number of equal segments you need, switch to Choose child subnet count. The count list only offers valid powers of two for the selected parent block, which keeps the split aligned with real prefix boundaries.

The first checkpoint is the summary strip and the three planner cards. Parent Block tells you whether the entered address normalized to the network you expected. Chosen Child Layout shows the first and last child CIDR values plus the gateway suggestion. Capacity Verdict tells you whether the chosen child subnet size meets the requested host target and, when it does, whether you are already at the deepest fitting split.

Split Planner is the comparison surface. Read it top to bottom when you want to see how a deeper prefix increases child count while shrinking usable hosts per child and raising the repeated reserved-address cost. Subnet Ledger is the handoff surface. Use it when you need exact network, usable range, gateway, and broadcast values for each child subnet.

The forecast chart is the quickest tradeoff view. The blue bars show usable hosts per child subnet. The green line shows how many child subnets each deeper prefix would create. When you enter a host target, the dashed red guide makes it obvious which prefixes fall below the minimum you need. That is often faster than reading every row in the planner table.

Pay attention to the range badge before you treat a result as deployable. A private RFC 1918 label fits ordinary internal planning. A documentation, benchmarking, link-local, loopback, multicast, reserved, or shared-address-space label means the arithmetic is still valid, but the operational meaning is different. The page helps you spot that difference early, before the wrong range spreads into downstream documents.

Step-by-Step Guide

Enter any IPv4 address inside the parent block and set Current CIDR to the parent prefix you want to split.
Choose whether you want to plan by Target CIDR or by Child subnet count.
Set the child size. The page will keep the two planning modes synchronized so the chosen prefix and child count always match.
If you care about host demand, enter Needed hosts per child. Leave it at 0 when you only want raw split arithmetic.
Open Advanced when you want to change the gateway host slot, shorten the visible ledger preview, or control CSV header and filename behavior.
Read the summary and planner cards first, then compare deeper prefixes in Split Planner. Switch to Split Forecast Chart for a faster visual tradeoff view.
Use Subnet Ledger when you need exact rows for documentation, then copy or download CSV, export DOCX, save chart images, or download JSON for the full calculation record.

Interpreting Results

The safest way to read the result is from the outside in. Start with the normalized parent block. If the parent network, wildcard, or usable span is already wrong, every child row will be wrong in a consistent way. Once the parent is correct, move to the chosen child layout and confirm the first and last child CIDR values. Those two edges tell you whether the borrowed bits and total sweep are behaving the way you expected.

Capacity Verdict answers a narrow but important question: does each child subnet fit the requested number of hosts? A positive figure means spare space remains in every child subnet. A negative figure means the selected child size is too small for that host target. When the page names a deeper prefix as the best fit, it is showing the most aggressive equal split that still leaves enough usable hosts per child.

How to read key IPv4 subnet split outputs
Result surface	Read it as	Common mistake to avoid
Parent Block	The real network boundary implied by the entered address and parent prefix	Assuming the typed host address is already the parent network
Chosen Child Layout	The selected child prefix, child count, gateway slot, and overall sweep	Checking only the count without checking the first and last child CIDR boundaries
Capacity Verdict	Per-child spare or short capacity against the requested host target	Treating a mathematically valid split as operationally sufficient for the real host load
Subnet Ledger	The exact handoff rows for implementation and documentation	Forgetting that the visible preview may be shorter than the exported ledger
Range type	A caution about the normalized parent block's address category	Reading `Public / other unreserved` as proof that the range is allocated to you or safe to advertise

Special-range labels need context. Private-use blocks are normal for internal planning. Shared address space belongs to carrier-grade NAT scenarios. Documentation and benchmarking blocks are useful for examples and labs, not production host pools. Link-local, loopback, multicast, and reserved ranges are stronger warning signs because they are not ordinary routed subnet choices. The page can flag those categories, but the final deployment decision still belongs to your routing, address-allocation, and policy review.

Worked Examples

One office `/24` turned into four equal VLAN blocks

Enter 192.168.20.41, set Current CIDR to /24, and keep the planning mode on Choose child CIDR with /26 selected. The page normalizes the parent to 192.168.20.0/24, shows four child subnets, and reports 62 usable hosts per child subnet.

If Needed hosts per child is 40, the verdict stays positive and the first child row becomes 192.168.20.0/26 with usable hosts 192.168.20.1 to 192.168.20.62. That is a clean fit for four similar VLANs without wasting time recalculating every boundary by hand.

Planning by child count instead of prefix

Suppose you know you need eight equal network slices from a parent 10.12.0.0/23 but you do not want to work out the child prefix first. Enter 10.12.0.18, set Current CIDR to /23, switch to Choose child subnet count, and choose 8 child subnets. The page resolves that to /26.

Each child subnet then carries 64 total addresses and 62 usable host slots. With a host target of 50, the split still fits comfortably. The planner table also makes it obvious that moving deeper than /26 would raise the child count while cutting usable space faster than the requirement allows.

A small block that exposes the `/30` boundary

Use a documentation-safe example such as 198.51.100.20 with Current CIDR set to /28, then choose /30 as the child prefix. The page will label the parent as a documentation block, create four child subnets, and report only 2 usable hosts per child subnet because the block size is four and two addresses remain reserved under the page's model.

If you ask for 3 hosts per child, the verdict goes negative and the page tells you the selected split is short. That is useful planning feedback. If you were actually sizing a point-to-point link, the right next step is not to force this page deeper. It is to switch to a workflow that models RFC 3021 /31 rules explicitly.

FAQ

Do I need to type the exact parent network address?

No. Any IPv4 address inside the parent block is enough. The page masks it back to the canonical parent network before it calculates the split.

Why does the child-count mode only show certain values?

Equal subnet splits must borrow whole prefix bits, so the valid child counts are powers of two relative to the selected parent prefix. The page only offers counts that map cleanly to real child CIDR values.

Why are /31 child subnets missing?

The visible workflow uses the conventional IPv4 host model with separate network and broadcast edges, so it stops at /30. RFC 3021 allows /31 on point-to-point links, but that special interpretation is outside this page's result model.

Can I use this for differently sized child networks?

No. Every child subnet in the result is the same size. If one segment needs a larger pool than another, you need a VLSM-oriented workflow instead of an equal-split planner.

Why did the detailed ledger disappear?

When the chosen split would create more than 4,096 child subnets, the page keeps the planner summary and chart but suppresses the detailed ledger to stay responsive.

Does the preview-row limit change what gets exported?

No. It only shortens the on-screen ledger. CSV, DOCX, and JSON still use the full rendered result set when the ledger is available.

What does Public / other unreserved mean?

It means the normalized parent did not match one of the built-in special-use categories. It does not prove the block belongs to you, is announced, or is acceptable under a provider or internal routing policy.

Does the calculation leave the browser?

No server-side subnet calculation is used here. The split, ledger, chart data, and JSON payload are produced in the page, then exported locally through the browser.

Glossary

Parent block: The original IPv4 network range that will be subdivided.
Child subnet: One equal-size network produced by borrowing additional prefix bits from the parent block.
Prefix length: The slash value such as /24 or /27 that states how many leading bits belong to the network portion.
Wildcard mask: The inverse of a subnet mask, often used in ACL and boundary reasoning to describe which bits may vary.
Usable host span: The assignable address interval between the network edge and the broadcast edge under the page's current rules.
Shared address space: The special IPv4 block 100.64.0.0/10, reserved for carrier-grade NAT environments rather than general private LAN planning.