A coworker of mine sent me an article written by a Cisco employee describing the IPv4 pool as likely to run out within 5 years. It’s titled “A Pragmatic Report on IPv4 Address Space Consumption. It certainly makes it sound like any new planning for networks should involve IPv6 capability.
DragonFly has IPv6 capability from KAME, which sounds like it will become much more useful very soon…
NAT has made IPv6 adoption less urgent. Theoretically, NAT can handle the load if all the existing IPv4 addresses are reallocated. Without NAT, IPv6 would have been needed years ago. But also, look at the source, Cisco. They can use IPv6 to force people to upgrade, so be skeptical of the claim for now.
Yeah, I thought the same thing when I read the article, as Cisco has a vested interest in getting people to upgrade.
However, the numbers do back up the situation, and I am noticing the IPv4 squeeze myself, at work.
Plus, if you want to be really cynical, the products required to handle large quantities of NAT sessions from Cisco are generally more expensive than having IPv6 support. :)
(a) Cisco is one of problems when it comes to IPv6 migration. The support simply sucks.
(b) NAT is evil and the issues with NAT will dramatically increase with the wide adoption of VoIP and similiar technologies. UPnP is not a solution.
(c) Areas like Mobile-IPv6 are still developing, but the need for proper roaming technologies is ever increasing.
(d) From the point of network administrators, IPv6 has the advance of allowing much simpler network topologies, thereby reducing costs.
(e) The most interesting reason for hackers to want IPv6 is the requirement to support many advanced and useful options like QoS, IPsec and IPcomp.