Upgrading the Internet – a Sobering Assessment

A report by a U.S. Department of Commerce Task Force concludes that transitioning the Internet from IPv4 to IPv6 will be much like all too many other IT upgrades âₓ long, tedious, expensive, doubtful in their ultimate benefits âₓ and ultimately unavoidable.

One of the major costs of society is maintaining infrastructure upon which it relies, whether it be in the public sector (e.g., roads and bridges) or the private (e.g., telecommunications). Maintenance is not elective, although the timing and cost of maintenance can at least be scheduled to balance annual costs, absent natural disasters.

Upgrading, on the other hand, can often be long-delayed, absent competitive pressures. Every private business, public agency, and utility needs to decide when the benefits of upgrading outweigh the costs, work and inconvenience of making the switch.

Not surprisingly, the Internet is no exception to this rule. Some years ago, the cost of building the infrastructure (routers, fiberoptic cable and so on) upon which the Internet is based was much in the news – as well as the question of whether the buildout of those resources could keep pace with exploding demand. Now, the infrastructure that supports the Internet and the Web, like the infrastructure of traditional utilities, is largely taken for granted and forgotten by those that depend upon it.

In the case of the Internet, however, some significant costs involve upgrading based not upon tangible goods, but upon virtual elements: the standards and protocols upon which the entire system depends. One of the most important of those protocols is the venerable and aptly named Internet Protocol (or IP), which enables not only the operation of the Internet itself, but many of the applications that run on top of it.

The IP is also a very complex standard, and upgrading the Internet to implement a new version of the IP will be very expensive. Currently, the Internet runs on IPv4 (and has done so for more than 20 years), even though IPv6 has already been developed.

One goal that is now forcing an upgrade to IPv6 is the belief by the U.S. government that upgrading to IPv6 will enhance cybersecurity. Accordingly, in preparation for the multi-year process of making the upgrade, a Task Force was established by a 2003 directive issued in a White House report, called the National Strategy to Secure Cyberspace. That Task Force was charged with determining what it would cost to make the transition, and with assessing the relationship between the cost and various anticipated benefits, such as augmenting national security and economic growth.

The complexity of making such an evaluation is increased by the fact that the Internet will not simply be switched over from supporting IPv4 to IPv6 some night at midnight, Greenwich Mean Time. Instead, there will be a dual-use period during which both versions of the IP will be in use. The Task Force was also therefore charged with thinking about how this protracted transition be accomplished without undue direct cost or undue negative economic impact.

Now, an 83 page report has been issued by that Task Force, with the title Technical and Economic Assessment of Internet Protocol Version 6 (IPv6). A related report of estimated transition costs, based on research by RTI International, has been issued by the National Institute of Science and Technology (NIST) as well.

The report reaches rather sober conclusions, noting (for example) that in the “initial years” security will not be augmented by the upgrade, and that in fact if the transition occurs too soon, security may actually suffer. At the same time, it anticipates that “most hardware, operating systems, and network-enabled software packages (e.g., databases, email, etc.) will likely include IPv6 capabilities within the next five years.”

The report does anticipate increased innovation as a result of greater availability of name spaces, and notes that at minimum the transition will “future proof” the Internet against address shortages. Other benefits âₓ none of them revolutionary – are described as well. At the same time, the report expects that full implementation will take many years, and that “Interoperability needs will be a major consideration in an enterprise’s decision to adopt IPv6.” Nonetheless, the report advises limiting costs by implementing IPv6 in the “normal refresh cycle” of upgrading hardware and software that currently supports IPv4.

The report is also sober in its evaluation of whether the upgraded network will indeed augment security, concluding that “new security paradigms that are significantly different than those commonly employed in today’s networks” will need to be developed at “considerable” expense in order to realize such benefits, and will be preceded by diminished, rather than increased, security. After 3 âₓ 5 years of implementation roll-out, the report concludes, users “may” see improvements in security.

Finally, the Task Force was charged with determining whether the government would need to play an active part in making the transition happen. The report concludes that no barriers will exist that would require that government intervene to assist the private sector, and therefore that “aggressive government action to accelerate deployment of IPv6 by the private sector is warranted at this time.”

Overall, one would conclude, upgrading the Internet will be much like all too many other IT upgrades âₓ long, tedious, expensive, doubtful in their ultimate benefits âₓ and ultimately unavoidable.


subscribe to the free Consortium Standards Bulletin (and remember to Buy Your Books at Biff’s)