Should You Care About IPv6?

As we move toward a world where web-based applications increasingly rely on both the public Internet as well as private intranets, it is important for collaboration managers to understand the technologies that underpin those networks, and what impact changes to IP networks may have on their application strategies.

IPv4, the current version of the Internet protocol, is almost 30 years old.  While it’s hard to argue with the success of IPv4, there are several problems which have led a number of people and organizations to push for an Internet-wide migration to a newer version of IP, IPv6.

Perhaps the biggest issue with IPv4 is exhaustion of available address space.  During the early years of IPv4 address space was easy to come by, and many organizations received far more space than they needed.  Since then address allocation procedures have become more efficient, but estimates from address registries such as ARIN (American Registry for Internet Numbers) suggests that available IPv4 address space will be exhausted sometime between now and 2012.  Demands for address space from emerging markets such as Asia, as well as increasing numbers of mobile devices continue to increase demand at a rate that exceeds supply.

Proponents of IPv6 point to its nearly unlimited address space (50 octillion addresses, that’s 5x10^28 for each of the 6.5 billion people on the earth) as its main benefit.  With all this address space network operators could eliminate their use of network address translation a tool for overcoming current address scarcity.  But IPv6 isn’t without problems of its own.  At this point there are no standardized approaches for enterprises to multi-home to multiple Internet service providers in an IPv6 environment.  Enterprises are also reluctant to move away from NAT, which is often seen as providing a security protection by hiding internal IP resources but also complicates applications such as VOIP and P2P services.  Finally, IPv6 products and services are still scarce, most Internet service providers still don’t offer IPv6 services, while IPv6 products in the security, application acceleration, and management spaces are still lacking.

Perhaps most importantly, adopting IPv6 requires adapting all current applications to support IPv6.  A recent test conducted by the University of New Hampshire InterOperability Lab found significant challenges in getting applications to work in an IPv6 environment. Given these challenges, the enterprise business case for migration to IPv6 simply doesn’t yet exist.  Nemertes Research has found little to no interest in migrating to IPv6 among our benchmark participants, but as IPv4 address exhaustion continues, enterprises will likely have to face the migration challenge sooner rather than later.

So what do collaboration managers need to know at this point?  First, be aware of IPv6 developments.  Ensure that any applications you buy today have a path to support IPv6 in the future.  But don’t panic, even if IPv4 addresses are exhausted in the next five years it could be another decade or more before enterprises need to consider moving their own infrastructures to IPv6 (assuming no other alternatives emerge).  Techniques such as NAT and address tunneling will continue to allow enterprises to survive the IPv6 transition with minimal disruption.