IPv4 and IPv6 addresses are the numeric identifiers that devices use to send and receive data on networks running the Internet Protocol (IP). Every device that talks on an IP network, including laptops, servers, phones, printers and cameras, has at least one. IPv4, the original version, uses short addresses that have effectively run out. IPv6 uses much longer addresses, providing a supply that is practically unlimited. Most networks today run both.
At a glance
- IPv4 addresses are 32 bits, written like 203.0.113.10, for about 4.3 billion in total.
- IPv6 addresses are 128 bits, written in hexadecimal like 2001:db8::1, for a practically unlimited supply.
- Addresses are grouped into prefixes, such as /24 in IPv4 or /64 in IPv6, that define network segments.
- Internal networks commonly use private IPv4 ranges and share a few public addresses through NAT.
- Public IPv4 addresses are scarce and cost money; IPv6 adoption is growing but uneven.
What problem it solves
Addresses are how networks know where to deliver data. A good address plan makes routing, security rules and troubleshooting straightforward: each site, department or function gets its own block, firewall policies can refer to those blocks, and anyone looking at an address can tell where it lives. A poor or undocumented plan causes overlapping ranges, broken connections between sites and painful renumbering when the business changes.
IPv6 exists to solve the IPv4 shortage. IPv4 was designed long before anyone imagined billions of connected devices, and workarounds such as NAT and carrier-grade NAT have stretched it. IPv6 removes the shortage so devices can have their own public addresses again and networks can be planned around structure rather than scarcity.
How it works
Structure. An address has a network part and a host part. The prefix length, written after a slash, says how many bits belong to the network. A /24 IPv4 network holds 256 addresses; a /64 IPv6 network, the standard size for one segment, holds about 18 quintillion.
Public and private. Public addresses are unique on the internet and are allocated by regional internet registries, such as ARIN in North America and RIPE NCC in Europe, to providers and organizations. Private IPv4 ranges (10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16) can be reused inside any network, with NAT translating to public addresses at the edge. Many providers also use carrier-grade NAT, which places customers behind shared public addresses.
Assignment. Devices get addresses either statically, typically for servers and network equipment, or automatically through DHCP. IPv6 devices can also configure their own addresses from information routers advertise. DNS maps names to addresses so people rarely need to type them.
Who owns them. Provider-assigned addresses come with your internet service and stay with the provider if you leave. Provider-independent addresses are registered to your organization, can be announced through more than one provider using BGP, and move with you.
Running both. Most networks are dual stack, running IPv4 and IPv6 side by side. Translation techniques such as NAT64 let IPv6-only devices reach IPv4-only services where needed.
For help with address planning and IPv6 rollout, see our IPv4/IPv6 Addresses solution page.
When it matters for buyers
- Mergers, acquisitions and divestitures. Combining or separating networks often exposes overlapping private ranges and shared public blocks.
- Changing internet providers. Provider-assigned addresses mean renumbering firewalls, VPNs, DNS and partner allow-lists.
- Adding public services or sites. Additional public IPv4 addresses may be rationed or charged for.
- Customer or regulatory requirements. Some contracts and public-sector buyers require IPv6 support.
- Cloud adoption. Cloud networks need address ranges that do not overlap with your offices and data centers.
Questions to ask vendors
- How many public IPv4 addresses are included, and what do additional blocks cost?
- Is IPv6 supported natively, and what size of IPv6 prefix will you assign us?
- Will our service sit behind carrier-grade NAT?
- Can we announce our own provider-independent addresses through your network?
- If we leave, how long do we keep your assigned addresses before they are reclaimed?
- Do your managed services, such as firewalls, VPN and monitoring, fully support IPv6?
How it differs from Internet Protocol (IP)
Internet Protocol (IP) is the protocol: the rules for packaging data into packets and forwarding them across networks. IPv4 and IPv6 addresses are the identifiers that protocol uses. Questions about routing and delivery belong to IP; questions about obtaining, planning and running out of addresses belong here.
Problems this helps with
- Need more public IP addresses and the provider won't give them
- Two merged companies use the same internal address ranges
- A customer or contract requires IPv6 support
- Changing providers means renumbering everything
