Around three years ago, I had a conversation with a very experienced network engineer about IPv6 and how it would improve the way we network. At the time, things were busy, but I managed to find some time to look into the technology. I was intrigued: A network protocol suite that offered a solution to our dwindling address base, tighter security measures, and features for streaming media applications would be fantastic. When would we have this miracle of modern science? "In a few years," seemed to be the general consensus.
Well, here we are, three years later, and we are still using the current version of TCP/IP, v4. Not having heard much about IPv6 recently, I set out to answer this question: "IPv6, where are we now?"
Ready, Set . . .
In 2001, the name of the game with IPv6 is integration rather than migration. Nearly all major operating systems have implementations of IPv6 available or currently in development. As you might expect, Unix and Linux platforms are particularly well catered for, although implementations for other platforms (including Windows, NetWare, and Macintosh) are also available. Networking hardware manufacturers are also approaching a state of complete readiness for IPv6; some major manufacturers already offer IPv6 support for their products, and others have beta offerings available.
So if the world is ready, the hardware is ready, and the software is ready, when can we expect to see IPv6 become part of our daily lives? It would seem that now it isn't the technology that's causing the delays, but rather the demand for the technology from the marketplace causing slower-than-expected progress.
|How Many Addresses Are a Lot?|
|As you may know, one of the biggest features that IPv6 will provide is a wider address space. Rather than using the 32-bit addressing format of the current version of TCP/IP, IPv6 uses a 128-bit addressing system. This system yields a staggering 340,282,366,920,938,463,463,374,607,431,768,211,456 addresses. Even though they are not all usable, they should keep us going--for a while, at least!|
Many companies aren't even asking themselves whether they need IPv6. One of the root causes is that many organizations don't understand IPv6. Or, rather, they don't see the need for it on their networks--and they may be right. Using Network Address Translation (NAT) to remove the requirement for large blocks of IP addresses has all but eliminated the need for many organizations to look for a solution to the lack of available IP addresses. In addition, other features of IPv6 are only useful to those organizations considering the deployment of IPv6-dependent applications, most of which are awaiting large-scale IPv6 adoptions before becoming affordable and commonplace. In the case of applications driving implementations, it is almost a case of the chicken and the egg.
Wireless May Lead the Way
Although IPv6 does offer more than just an expanded address base, this feature is seen as the biggest driver behind a wide-scale implementation of the new protocol. The largest use of the expanded address base is that of mobile wireless. Even though in certain areas of the world the adoption of wireless devices has been slower than was hoped, the growth potential for this market area is staggering--it will truly need the addressing and other features of IPv6.
An interesting tilt on these statistics was made during a presentation by Nokia at a recent IPv6 forum meeting. The company stated that at present, China has the use of only 4 million IPv4 addresses--but the country currently has over 50 million cellular handsets. If all of these handsets needed an IP address, the only practical solution would be to use IPv6. And mobile handsets are only the beginning: Consider that every home, office, factory, hotel room, and hospital room will be loaded with wireless IP devices in the future, and the need for an almost self-configuring protocol with sufficient addresses becomes even more evident.