IPv6 cost estimates trigger debate: Part 1

How much will it cost ISPs and enterprise customers to transition to IPv6, an upgrade to the Internet’s main communications protocol? No one knows for sure, but a recent report sponsored by the U.S. federal government has created controversy among IPv6 proponents for its $25 billion estimate.

RTI International asserts that the transition to IPv6 will cost $25 billion over the next 25 years, primarily to cover increased labor costs including training. Two arms of the U.S. Commerce Department – the National Institute of Standards and Technology (NIST) and the National Telecommunication and Information Agency (NTIA) – funded the RTI study.

“Although these cost estimates seem large, they are actually small relative to the overall expected expenditures on IT hardware and software,” Michael Gallagher, a senior researcher at RTI International who directed the study, said in a statement. “The costs are even smaller relative to the expected value of potential market applications of the IPv6 version.”

Users will incur the majority of the IPv6 transition costs, according to RTI. RTI breaks down the $25 billion cost as follows:

* Users: $23.3 billion

* Infrastructure vendors: $1.4 billion

* Application vendors: $.6 billion

* ISPs: $.1 billion

* Total Cost: $25.4 billion

RTI says the hardware and software costs to upgrade to IPv6 will be negligible because the new protocol will be built into next-generation hardware, operating systems and software that are purchases as part of standard upgrade cycles. Instead, the primary cost will be in training and related labor expenses.

The RTI study says it is difficult to quantify the potential benefits of IPv6 because the protocol has not been deployed in the U.S. and new applications that take advantage of its special features have not yet been developed. The report estimates that the potential benefits of IPv6 will exceed $10 billion per year, far outstripping his estimate of transition costs of $1 billion per year.

RTI says that over time the benefits of IPv6 should “greatly exceed” the transition costs of $25 billion, which will be spread over 25 years. These benefits include VoIP, remote access products and services and improved network operating efficiencies, RTI says.

IPv6 advocates say the $25 billion number is too large.

“We think the real number is much lower, but we don’t want to guess what it is,” says Jim Bound, chair of the North American IPv6 Task Force and a senior fellow with HP. “We don’t believe the estimates for the return on investment. We think those numbers are going to be higher.”

Bound says the report should have focused more on the cost to U.S. ISPs and businesses of not doing IPv6, which is being deployed in Asia and Europe at a much faster rate than it is in the U.S.

“What is the cost of not getting to the next generation network? What is the cost of not returning to an end-to-end model?” Bound asked. “We think more work needs to be done to quantify all the benefits you get from IPv6.”

The RTI study was designed to provide a cost-benefit analysis of the conversion to IPv6. The RTI study comes on the heels of last year’s mandate by the Office of Management and Budget that all federal agencies must run IPv6 on their network backbones by June 2008. The Defense Department issued a similar mandate in 2003 to have all military networks transitioned to IPv6 by FY 2008.

Developed by the IETF, IPv6 promises easier administration, tighter security and an enhanced addressing scheme when compared to IPv4, the Internet’s current protocol. IPv6, which uses a 128-bit addressing scheme, supports a virtually limitless number of uniquely identified systems on the ‘Net, while IPv4 uses a 32-bit addressing scheme and supports only a few billion systems.

More on the RTI Study can be found here.

Coming up in the next few issues of the Service Provider News Report, we’ll look more closely at RTI’s IPv6 transition cost estimates for ISPs and enterprise customers as well as the benefits of IPv6.