Nothing But Net
Mark McFadden Addressing the Need for More Addresses
- By Scott Bekker
- September 07, 2000
Can you remember when people were worried that we would run out of addresses for Internet devices? Looking at the pace of growth of the Internet some people projected that we would run out of IPv4 address space as early as 2006.
Since that time, two important measures were undertaken. First, a stern rationing program for Internet addresses was implemented. The combination of draconian IP address space management along with the transition from class oriented address allocation (remember Class A, Class B and Class C allocations?) to Classless Interdomain Routing (CIDR) slowed the growth in IP address utilization. The Internet registries have been successful in ensuring the conservation of a scarce resource -- addresses.
The second step was the development of a replacement for the twenty-year-old Internet Protocol (IP). The current version, IP version 4, has a built-in limitation to the number of addresses available. The Internet engineering community seized on the opportunity to develop a new IP protocol, called IPv6 with a substantially larger available address space. For comparison IPv4 has a 32-bit address space that provides about seven billion addresses -- incredibly, IPv6 provides 340,282,366,920,938,463,463,374,607,431,768,211,456 usable IP addresses.
Address space is not the only motivation for IPv6. Other features of IPv6 address critical business requirements for more scalable network architectures, improved security and data integrity, integrated quality-of-service (QoS), autoconfiguration, mobile computing, data multicasting, and more efficient network route aggregation at the Internet’s backbone. IPv6 is a big package, and addressing is only the most visible component of the work.
What’s remarkable is that next to nothing is happening in North America with IPv6.
You would expect the nation that leads the world in Internet connectivity, bandwidth, and business applications to be on the leading edge of such a critical technology. But in fact, North America is far behind in the transition to IPv6 and its advantages.
The Regional Internet Registries (RIRs) are non-profit organizations that dole out resources such as IP address allocations to large service providers and telecommunications providers. Europe’s RIR, Reseaux IP Europeans, has given out more than twenty-two allocations of IPv6 address space to European network operators. In the Asian Pacific region, the Asian Pacific Network Information Center has assigned more than fifteen. In North America the number is a paltry six.
Why the lack of interest in a key technology for the future of the Internet?
Part of the reason must be in Network Address Translation (NAT) boxes. I recently reviewed Microsoft’s beta of Internet Security and Acceleration Server for the print version of ENT. The ease with which NAT services can be implemented and configured is astonishing. No wonder use of address space has slowed -- many companies are simply "hiding" their networks using NATs and private addressing.
Also, demand for IPv6’s advanced services -- like QoS and autoconfiguration -- seems pretty limited.
Still, the lack of interest is surprising. Given the potential explosion of mobile computing -- and the addressing demands it will place on us -- it seems certain there will be a surge in addressing requirements. When that time comes will your organization be in a position to take advantage of IPv6? Or, like so many organizations today, will you continue to patch and cope with a twenty-year-old protocol? --Mark McFadden is a consultant and is communications director for the Commercial Internet eXchange (Washington). Contact him at email@example.com.
Scott Bekker is editor in chief of Redmond Channel Partner magazine.