The History of IPv6
The history of IPv6 is the story of the internet seeing its own future and quietly building an escape hatch. This guide traces how engineers realised IPv4 would run dry and designed a vastly larger addressing system to take its place.
Why the internet needed IPv6
The history of IPv6 begins with a simple counting problem. The original Internet Protocol, IPv4, gives every device a 32-bit address, which allows for about 4.3 billion unique addresses. In the early days that felt like an endless supply. But as the internet spread from a handful of research institutions to homes, offices, and eventually pockets around the world, engineers began to see that four billion addresses would not stretch far enough.
By the early 1990s the growth of the internet was steep enough that the technical community grew genuinely worried. Every new network, every new computer, and eventually every new phone would need an address. The mathematics were unforgiving: at some point the pool would empty. The response was to design a successor with so much room that the problem would never return. You can read more about the crunch it was built to solve in our guide to IPv4 address exhaustion.
What makes this foresight remarkable is the timing. The warning came while IPv4 addresses were still comfortably available, at a moment when many people would have said there was no problem at all. Acting on a shortage years before it bites requires a kind of institutional patience that is rare in technology, where the pressure is almost always to solve today's problem and leave tomorrow's for later. The internet's engineers chose the harder path, and the modern web is better for it.
IPv6 was designed decades before it was truly needed. The internet effectively planned its own escape from running out of room long before the last IPv4 addresses were handed out.
Designing a bigger internet
The engineering community that guides internet standards set out to design a next-generation protocol. Several proposals were considered before the group settled on the design that became IPv6. The headline decision was the address size: instead of 32 bits, IPv6 uses 128 bits.
That change sounds modest but the effect is staggering. Each extra bit doubles the number of possible addresses, so moving from 32 bits to 128 bits does not add addresses, it multiplies them almost beyond comprehension. IPv6 offers roughly 340 undecillion addresses, a number with 39 digits. For any practical purpose it is inexhaustible. The designers wanted to make certain that no one would ever have to repeat the exercise.
It is hard to picture a number that large, so consider the scale another way. IPv6 provides enough addresses to assign many trillions of them to every single person alive, with a practically limitless surplus left over. The point was never to allocate them efficiently in the way IPv4 forced everyone to. The point was to have so much room that address scarcity would simply cease to be something engineers ever had to think about again.
What IPv6 changed beyond size
Address space was the headline, but the designers took the opportunity to modernise other things too. IPv6 simplified the structure of the packet header so routers could process traffic more efficiently. It built in support for automatic address configuration, letting a device generate its own address on a network without always needing a separate server to hand one out. It also made features like large-scale address allocation and network renumbering easier to manage.
There was another philosophical shift buried in the design. IPv4's scarcity had forced the internet to lean heavily on address-sharing tricks, which meant many devices no longer had a truly unique, reachable address of their own. IPv6 restored the original vision of the internet, in which any device could, in principle, address any other directly. That end-to-end ideal had been part of the network's founding design, a story we touch on in our history of TCP/IP, and IPv6 was in many ways an effort to bring it back.
| IPv4 | IPv6 | |
|---|---|---|
| Address size | 32 bits | 128 bits |
| Total addresses | About 4.3 billion | About 340 undecillion |
| Example | 192.0.2.1 | 2001:db8::1 |
| Notation | Decimal, dot-separated | Hexadecimal, colon-separated |
| Auto-configuration | Needs help | Built in |
If you want a fuller side-by-side, our guide on IPv4 vs IPv6 walks through the formats and trade-offs in detail.
The long, slow rollout
Here is the twist in the story. IPv6 was ready as a standard long before the world actually needed to use it, and yet adoption crawled. The reason is that IPv4 and IPv6 are not directly compatible, so the internet could not simply flip a switch. Every network, operating system, router, and service had to learn to speak the new protocol, and there was little pressure to hurry while IPv4 addresses were still available.
To buy time, the internet leaned heavily on workarounds. The most important was Network Address Translation, which lets many devices share a single public IPv4 address. NAT was so effective at stretching the old system that it removed much of the urgency to migrate. The result was a decades-long transition in which the two protocols run side by side, a story we explore further in why IPv6 adoption has been so slow.
Where IPv6 stands now
Despite the slow start, IPv6 is now a real and growing part of the internet. A large and rising share of traffic to major services travels over IPv6, mobile networks have embraced it widely because they connect so many devices, and most modern operating systems and home routers support it out of the box. Your own connection may well be using IPv6 already without you noticing.
Much of the transition has happened invisibly, which is exactly how good infrastructure is supposed to work. Modern devices often run both protocols side by side and quietly pick whichever one succeeds for a given connection, an approach commonly called dual stack. The upshot is that ordinary users rarely have to think about which version they are using. The internet handles the choice for them, and the enormous machinery of the changeover stays hidden behind the scenes.
The transition is not finished, and IPv4 is not gone. For the foreseeable future the two will coexist, with software quietly choosing whichever protocol works for a given connection. But the long arc of the story is clear: the internet foresaw its own limits and built the road out well in advance.
The lesson of IPv6
It is also a reminder that the internet's biggest problems are often solved quietly, long before the public ever hears of them. The people who allocate addresses, as our guide to who runs the internet describes, cannot afford to wait until a crisis arrives. IPv6 is what careful stewardship looks like in practice: a decades-long, largely invisible effort to make sure the network you use every day never simply fills up and stops growing. Few users will ever notice the work, and that quiet success is precisely the point.
The history of IPv6 is a rare example of long-term planning in a fast-moving field. The people who built it were not solving a problem they had; they were solving one they could see coming years away. That foresight is why the modern internet, with its billions of phones, sensors, and connected gadgets, has somewhere to grow. To see how it fits into the wider story, visit our brief history of the internet, or head back to IP Animals to check whether your own connection is already speaking IPv6.
Frequently asked questions
Why was IPv6 created?
IPv6 was created because the older IPv4 system offers only about 4.3 billion addresses, which engineers realised would not be enough for a growing global internet. IPv6 uses much larger 128-bit addresses so the supply is effectively unlimited for any foreseeable future.
How many addresses does IPv6 have?
IPv6 uses 128-bit addresses, giving roughly 340 undecillion possible addresses, written as 2 to the power of 128. That is an astronomically large number, far more than IPv4's 32-bit space of about 4.3 billion.
Is IPv6 replacing IPv4?
Gradually. IPv6 was designed as the long-term successor to IPv4, and adoption has grown steadily, but the two run side by side today. Most of the internet still supports IPv4 as well, and the transition has taken far longer than its designers hoped.
What does an IPv6 address look like?
An IPv6 address is written as eight groups of hexadecimal digits separated by colons, such as 2001:db8::1. Long runs of zeros can be shortened with a double colon, which is why many IPv6 addresses look much shorter than the full form.