Is there really a need for IPv6?

For some time now, ARIN has been pushing providers to switch to IPv6. The steady rise of computer use along with the proliferation of other web devices like cell phones and PDAs has put a heavy burden on the IPv4 system, which is expected to run out of addresses by 2011. Not so, says Paul Rubens in a recent Enterprise Networking Article.

Rubens states that most network providers have been too lazy to upgrade to IPv6, but argues there is no reason for them to spend the money on the equipment to do so. There are numerous changes that could be made to the antiquated IPv4 system to further its lifespan, including freeing up hundreds of millions of IPs reserved but not being used and using Network Address Translation.

NAT allows multiple web devices to share an IP address. The author estimates that by sharing an IP address with two web users, IPv4 would last for another 20 years. If that number were increased to 10, then a new system would not be needed for another century. In addition, there are still 925 million IPv4 addresses left that have yet to be allocated.

IPv6 is able to hold 3.4×10³⁸ addresses, compared to IPv4’s measly 4.3 billion. To put that into perspective, the new system has room for 7.9×10²⁷ more web devices, or 5×10²⁸ addresses for every individual on the planet.

There is essentially no limit to IPv6. While Rubens is right in that we could optimize our use of IPv4 and make it last longer, many providers have already made the switch. Sure, upgrading hardware for the new system will be costly, but so will implementing work-arounds to extend the life of IPv4. Finally, given the fact that the IP system is necessary for the Internet to function, do we really want to take chances with it? Better safe than sorry and IPv6 ready, I say.