The Road to IPv6 Is Longer Than Expected
August 7, 2013 Leave a comment
- IPv6 migration worldwide is dragging along at a glacial pace. Yet, it is no wonder enterprises are not planning on migrations anytime soon.
- Enterprises will be deploying IPv6 transition strategies long before they will be migrating their internal networks to IPv6.
Whither IPv6? When the Internet Assigned Numbers Authority (IANA) allocated the last IPv4 address block in January 2011, the prognosis did not look good for IPv4. Shortly thereafter, some of the regional Internet authorities stopped allocating IPv4 blocks downstream, reserving a small portion for IPv4-to-IPv6 transition strategies. Here in the U.S., Comcast and Time Warner started rolling out IPv6 to business and residential customers while Verizon Wireless began using IPv6 on its 3G/4G networks.
According to statistics gathered by Google showing the percentage of visitors coming from IPv6 networks, the countries with the highest adoption are Switzerland with 10.21% and Romania with 8.31%. Adoption in the U.S. is only 3.52%. Only 1.75% of Google users worldwide are coming from IPv6, though the vast majority are coming in from a native address versus a tunnel.
For some, the slow adoption of IPv6 overall is surprising. It is available on all modern desktop, server and mobile operating systems as well as switches and routers. Network-attached appliances such as printers, cameras, VoIP phones and so on are lagging along, which is a significant concern for enterprise IT as well as consumer broadband providers, but there are strategies to address those situations. When I talk to enterprise network administrators, however, IPv6 migration is still pretty low on the priority list.
The panic over IPv4 exhaustion has subsided as everything continues to hum along. The fact is, IPv4 works inside and outside the enterprise, and there is not much of reason to migrate as long as it continues to do so. The power of maintaining the status quo – sticking with IPv4 and using NAT – has great stickiness, and until there is a compelling reason to migrate, enterprises will not expend the effort.
What I think will drive the change is when consumers and employees outside the enterprise on IPv6 networks fail to reach Internet servers with IPv4 addresses. The cost for customer support on the consumer side and help desk calls on the employee side will rise enough to make it cheaper to switch than to fight. Even then, I think IPv6 migration within enterprises will take longer than anyone expects, and because of that we’ll see the importance of IPv4-to-IPv6 conversion dramatically rise.