The problem it solves
Every device that talks directly on the internet needs an address. The original system, IPv4, uses 32-bit addresses, which allow about 4.3 billion unique combinations (2 to the 32nd power is 4,294,967,296). That sounded enormous when the design was settled in the early 1980s, when the network was small and mostly academic.
Then came personal computers, the web, smartphones and billions of connected gadgets. Addresses were also handed out in blocks that were not always used efficiently, and many were reserved for special purposes. The pool of unassigned IPv4 blocks held by the central registry ran out in 2011, and the regional registries that distribute them have since reached or approached depletion at different times. Today, new IPv4 addresses are generally obtained by transfer from others, often at a cost.
The stopgap: sharing addresses
Even before the shortage became acute, engineers worked on a longer-term fix, and the internet also found workarounds. The best known is network address translation, in which an entire household or company sits behind one public address, with private addresses used inside. Mobile carriers and some providers go further by sharing a public address among many customers.
These workarounds extended IPv4’s life, but they add complexity. Sharing makes it harder to host services at home, complicates some applications such as video calls and games, and can make troubleshooting and security investigations harder. The article on IP addresses explains how the basics of NAT work.
What IPv6 changes
IPv6 uses 128-bit addresses. The number of combinations is 2 to the 128th power, roughly 3.4 followed by 38 zeros, or about 340 undecillion. It is so large that it is not practical to compare it with the number of devices. The aim was not just to have more addresses but to make address shortage a non-issue for the foreseeable future, so that every device can have its own globally unique address if desired.
An IPv6 address is written as eight groups of four hexadecimal digits, separated by colons. An example from the range reserved for documentation:
2001:0db8:0000:0000:0000:ff00:0042:8329
There are two shortening rules: leading zeros in a group can be dropped, and one run of consecutive all-zero groups can be replaced with a double colon. The example above becomes 2001:db8::ff00:42:8329. The double colon may appear only once in an address, or it would be ambiguous.
Other design differences
- Simpler headers: the fixed part of each packet’s header was streamlined, which can make processing by routers more efficient.
- Automatic configuration: devices can often configure their own addresses by combining a network prefix announced by the router with their own identifier, without needing DHCP, although DHCP for IPv6 also exists.
- Multiple addresses per device: it is normal for an interface to have several addresses at once, such as one for local use and one for the wider internet.
- No broadcasts: IPv6 relies on multicast instead of the broadcast messages used in IPv4, which is meant to reduce unnecessary traffic.
- IPsec support: the specification includes support for IP-level encryption, although it is optional in practice, and most web security continues to rely on HTTPS encryption at a higher level.
Is IPv6 faster or more secure?
Not automatically. Some people report better performance in particular cases because it can avoid layers of address translation, but the difference varies and is often small. For security, removing the need for NAT does not remove the need for a firewall. NAT is not a security feature by design, but devices behind it were not directly reachable by default, and IPv6 networks need an explicit firewall policy to achieve similar protection. Home routers that support IPv6 generally ship with sensible defaults, though it is worth confirming.
Privacy is another point. Early IPv6 addressing could embed a device’s hardware identifier into its address, which made it trackable across networks. Modern systems commonly use temporary, randomized addresses for outgoing connections to reduce this.
Why the transition is slow
IPv6 and IPv4 are not directly compatible: a device that speaks only IPv4 cannot talk straight to one that speaks only IPv6. Because of this, the transition has been gradual, with most networks running both at once, an approach called dual stack. Other techniques translate or tunnel IPv6 traffic across IPv4 networks.
Moving is costly and, for many organizations, there is little immediate payoff, since workarounds still function. Adoption therefore varies a great deal by country and provider. Some regions and mobile networks use IPv6 for a large share of traffic, while others lag behind. Public statistics tracked by large network operators show the share changing over time, so check a current source for exact figures.
What this means for ordinary users
Most people do not need to do anything. Operating systems, browsers and many home routers already support IPv6, and if your provider offers it, your traffic may use it without you noticing. Name lookups return a record type for IPv6 addresses, as covered in the guide to the domain name system, and your device picks whichever works best.
If you are curious, you can check your connection with an online test that reports whether you have an IPv6 address. Not having one is not a fault. If you manage a home network and experience odd problems, such as slow page loads, a service may be offering IPv6 that works poorly on your line, and temporarily disabling IPv6 on the device is a diagnostic option, although fixing the cause is better.
The longer view
IPv6 is best understood as the long-term replacement for an addressing scheme that was designed for a far smaller network. The two systems will likely coexist for many years, and the exact date when IPv4 fades is unknown. What is clear is that the number of connected devices, from phones to IoT devices, continues to rise, and a roomy address space is what lets that growth continue without further patchwork.