IPv6 Expert Silvia Hagen says ‘IPv4 is History’

There is a saying that goes like this, “There is only one thing mankind can learn from history and that is that mankind never learns from history.”

But what does history show? History shows that for instance in the early 1980s leading management schools performed in-depth analysis to find out if it was feasible that someday a computer would be on every desk. They came to the conclusion that this did not make sense and would never happen. But only a few years later many employees had more than one computer on their desk and one at home.

So what about IPv6? IPv6 is the successor protocol to IPv4. IPv4 was developed in the early 1970s by inspired engineers at American universities. There was no business plan for IPv4 and no business plan for the Internet. The engineers simply wanted to create a protocol to connect a bunch of computers meaning they didn’t have a design goal for creating the Internet. Nobody would have funded it anyway.

In the early 1990s some people started using the “Internet”. At that time it was mostly CompuServe users engaging with each other in CompuServe Forums. The World Wide Web was born in 1990, when Tim Berners-Lee set up the first successful communication between a browser and a web server at CERN in Switzerland. Therefore we can say that the Internet we know today really started in the second half of the 1990s. And it didn’t take long until many people could not imagine living without it. How do you want to find information, buy books, compare prices, book flights, learn about new topics and and and…if you can’t use the Internet?

So why am I pointing this out in an article about IPv6? Simple: IPv6 is about the continuation of the Internet, nothing more and nothing less. There is no specific business case, just like IPv4 and the Internet. But there is also no way around IPv6 either, because it is time for the next evolutionary step of the Internet. Today we are down to 5% of available IPv4 addresses[1]. This equals 14 /8 blocks (a /8 block is a Class A address block). With an average consumption of approximately 10/8 per year in the last couple of years (and 10/8 in the first 6 months of 2010), the pool of available addresses will finally be exhausted around May or June 2011, and maybe even earlier. The Internet has therefore already started to change to IPv6.

Although only about 25% of the world population has Internet access, you can expect the growth rate to remain stable or to even increase. Not to mention new services and technologies that are developed in different industries, such as sensor and monitoring systems for devices of all types and buildings which require permanent IP addresses. The networked car of the future, in development at most of the major car vendors, will need up to 50 IP addresses (how many cars are out there?) and in the not so distant future global telephone communication will be IP-based – go figure. 

Only IPv6 can cover this exponentially growing demand for IP addresses. And it doesn’t only solve the address issue, it has been optimized for the high and growing demands of our complex networks and will be able to scale, where IPv4 can never go. As the Internet changes, enterprises must also adopt IPv6.

Do you want a glimpse on how vast the IPv6 address space is? Here we go. By January 2010 approximately 140’000 /32 have been allocated. A /32 is what is used to count IPv6 allocation. A /32 leaves 32 bits open for subnetting and is usually allocated to ISPs by their regional registry. So we can say that ONE /32 is more address space than the whole IPv4 address space (because an IPv4 address has 32 bits in total). So we have already given out 140,000 times the IPv4 address space[2]. How much do you think this is in percentage of the total available address space? It equals 0.026% of the currently defined global unicast space. Wow!

So because this is not history, we may be able to learn from these facts. I for my part learned that a) there is no way around IPv6 and b) the wise will start introducing it while there is time and they can do it gradually, learning as they go. Many vendors are waking up and developing and optimizing their products for integration in IPv6 networks and for support of dual-stack networks. Companies in the IPAM space such as BlueCat Networks  offer solutions like Proteus which enables enterprises to harness IPv6 deployment.

Some will choose to wait until the last minute, hiding their heads in the sand, pretending there is no need for IPv6 and that there is no business case either – and some day they will wake up and realize they missed something, they missed opportunities, they missed markets, they missed doing it the easy way.

---

[1] http://www.potaroo.net/tools/ipv4/

[2] http://www.bgpexpert.com/addrspace-ipv6.php