We ran out Of internet address space in 1974

During his keynote on Thursday at OSCON Peter Salus reminded us that many of the stories that we discuss today about technology are just replayed from years past, the names change and the scale grows but the stories are fundamentally the same.

Increasingly I find myself involved in discussions about IPv6 implementation, it was discussed at the OECD Ministerial on The Future of The Internet two months ago in Korea, and it will be central to discussions at the next IGF forum in India in December.

One of Peter’s stories related to a problem that was dealt with in the early days of the ARPAnet, when they ran out of network address space in 1974. Sound familiar?

A quick search unearthed a fascinating write up on the origins of the Internet on NetValley.com, it is labeled as a story told to Bernard Aboba by Vint Cerf.

The paragraph in the story that supports Dr. Salus’ comment reads;

The original design of 1973 and 1974 contemplated a total of 256 networks. There was only one LAN at PARC, and all the other networks were regional or nationwide networks. We didn’t think there would be more than 256 research networks involved. When it became clear there would be a lot of local area networks, we invented the concept of Class A, B, and C addresses. In Class C there were several million network IDs. But the problem that was not foreseen was that the routing protocols and Internet topology were not well suited for handling an extremely large number of network IDs. So people preferred to use Class B and subnetting instead. We have a rather sparsely allocated address space in the current Internet design, with Class B allocated to excess and Class A and C allocated only lightly.

The history raises some interesting points, especially as we are now talking more and more about the need to roll out IPv6.

Back in 1974 there were many fewer nodes on the internet, and all of them run by research and academic organizations. The flaw in the original design was something that could be fixed quickly across the limited number of networks as the idea of class A, B, and C addresses was adopted and implemented in the way that Dr. Cerf describes.

Clearly the Internet today is significantly more complex, involving a diverse array of software and hardware from many hundreds of thousands of providers from around the globe.

Rolling forwards in time a little to 1998 I found this story, originally from WindowsITPro. The story reminds us that there was a strong belief that we would run out of Class B addresses in March of 1994, but through the invention of standards for network address translation and other address management tricks we were able to push that date out to somewhere between 2005 and 2010. The story also points us towards RFC 1792, a specification that was originally published in January 1995 detailing what we know as IPv6.

For what it is worth, the software and hardware industry are pretty close to being ready for implementation, you can follow Microsoft’s progress here.

In the Internet centric world that we live in today the implementation of IPv6 not only relieves pressures that we feel on available address space but it also makes a whole host of new scenarios possible as the ability to grant an address to every device on the planet becomes a reality.

We all have a role to play in pushing forwards with IPv6 implementation. Here are a few things you can personally do to do your part and get yourself ready;

First learn a little about IPv6 itself, there are plenty of sites that provide information and training. Here is one example, or Microsoft’s own primer which is here.

Second you can work out what is going on in your country around IPv6 backbone support. Almost every country around the region has at least a grass roots effort to drive IPv6 adoption. Here are examples from Australia, India and the Pacific Islands.

Third, learn a little about technologies like Teredo Tunneling, which provides mechanisms for tunneling IPv6 traffic over existing IPv4 networks.

Fourth, keep track of the growing number of operating systems, applications and pieces of hardware that are IPv6 compliant.

Finally, think about the applications where IPv6 services would help your organization recognize more value of our your networked assets and look for opportunities to use the technology.

Moving beyond the network address issues of 1974 took a little creative work by the entire internet community, moving from IPv4 to IPv6 will involve the same level of participation – but this time from an exponentially larger community!

We all have a part to play, they managed the change last time, we can do it again.