IP Calculator

Subnetting, in plain terms

A CIDR prefix like /24 tells you how many of the 32 bits in an IPv4 address are fixed as the "network" portion — the rest are free for host addresses. A smaller number after the slash means a bigger network; a bigger number means a smaller one. /24 gives you 256 addresses, /25 gives you 128, and so on, halving each time you add a bit.

Every subnet reserves two addresses you can't assign to a device: the network address (all host bits zero — identifies the subnet itself) and the broadcast address (all host bits one — reaches every host on that subnet). That's why usable hosts is always total addresses minus 2, except for point-to-point links (/31, which skips the reservation entirely per RFC 3021) and single-host routes (/32).

VLSM — sizing subnets to fit, not force-fitting one size

Variable Length Subnet Masking just means: don't give every subnet the same size when they don't need the same number of hosts. A 2-host point-to-point WAN link doesn't need a /24; a /30 wastes far less space.

The standard approach — and what this tool does — is to sort your requirements largest-first and allocate from the top of your address block down, aligning each subnet to a boundary that matches its own size. Allocating largest-first avoids fragmentation: if you allocated smallest-first, you could end up with leftover gaps too small for a later, bigger requirement.

Supernetting & route summarization

Supernetting (route aggregation) is the reverse move: instead of splitting a block, you're rolling several contiguous networks up into one larger advertised route, to keep routing tables small. 192.168.0.0/24 through 192.168.3.0/24 aggregate cleanly into 192.168.0.0/22 because they're contiguous and the count is a power of two.

Watch for "leaky" aggregation — the tool will flag it. If your networks don't align to a clean power-of-two boundary, the smallest covering block will include address ranges you didn't actually list, meaning your summary route silently advertises reachability to space that isn't really behind that route.

IP address classes (classful addressing, mostly historical)

Before CIDR existed, IPv4 space was divided into fixed-size "classes" based on the first few bits of the address. Routing today is entirely classless — VLSM and CIDR make the class system obsolete for actual network design — but the ranges still show up constantly in documentation, exam questions, and the reserved blocks below, so it's worth knowing.

Note the gap: 127.x.x.x is carved out of the old Class A range entirely for loopback (127.0.0.1 — "localhost"), not for host addressing.

Public vs. private IP ranges

Private ranges (RFC 1918) are reserved for use inside local networks and are never routed on the public internet — every home router, office LAN, and data center uses addresses from these blocks internally, then relies on NAT to translate to a public address at the edge.

Everything outside those three blocks (and outside the other reserved ranges below) is public — globally routable and, if you announce it, reachable from anywhere on the internet. A few other special-purpose ranges worth knowing:

• 127.0.0.0/8 — loopback, always refers back to the local machine.
• 169.254.0.0/16 — link-local (APIPA), auto-assigned when DHCP fails; means "no address was actually configured."
• 100.64.0.0/10 — shared address space (RFC 6598), used by ISPs for large-scale NAT (CGNAT) without touching RFC 1918 space.
• 224.0.0.0/4 — multicast, corresponds to the old Class D range above.

IPv6 quickstart

IPv6 addresses are 128 bits, written as eight groups of four hex digits, with :: allowed once to compress a run of zero groups. Prefix conventions are different from IPv4 habits: ISPs typically hand out a /32 to /48 per customer, and a /64 is the standard size for a single LAN segment — not because you need that many hosts, but because stateless address autoconfiguration (SLAAC) is built around a 64-bit interface identifier.

Because address space is so large, "how many hosts fit" is rarely the question with IPv6 — it's almost always answered "plenty." Planning is more about how you divide space hierarchically (site → building → floor → VLAN) than about conserving addresses.

Frequently asked questions

What does a CIDR prefix like /24 mean?
A CIDR prefix such as /24 specifies how many of the 32 bits in an IPv4 address are fixed as the network portion. The remaining bits are available for host addresses, so /24 yields 256 addresses, /25 yields 128, and so on.

What is VLSM (Variable Length Subnet Masking)?
VLSM lets you size each subnet to match its actual host requirement instead of using one fixed size everywhere, so a 2-host link can use a /30 while a 50-host LAN uses a /26 from the same address block.

What is supernetting or route summarization?
Supernetting combines several contiguous networks into one larger advertised route to keep routing tables smaller, provided the networks align to a clean power-of-two boundary.

What is the difference between a public and private IP address?
Private IP addresses (RFC 1918 ranges 10.0.0.0/8, 172.16.0.0/12, and 192.168.0.0/16) are reserved for internal networks and are never routed on the public internet. Any address outside these reserved ranges is public and globally routable.

What are IP address classes A, B, and C?
Classful addressing divided IPv4 space by the first octet: Class A (1–126) with a default /8 mask, Class B (128–191) with /16, and Class C (192–223) with /24. Modern networks use classless CIDR addressing instead, but the class ranges still appear in documentation and reserved blocks.