Running on Empty: The Challenge of Managing Internet Addresses

Abstract

Over a quarter century has passed since IP addresses from the Internet’s now globally ubiquitous (IPv4) address pool first began to be distributed to and used by operators of the 28,000 and counting independent routing domains that constitute today’s Internet. At current demand rates, the total pool of IPv4 addresses is expected to be fully allocated within the next 3-5 years. Despite the looming exhaustion of the address pool, many Internet network operators have exhibited reluctance to transition to the successor address resource pool (IPv6). In light of the looming scarcity of IPv4 addresses, there is growing interest in the potential for market-based allocation mechanisms to induce more efficient use of the IPv4 address space and to smooth the transition to IPv6. Some of these proposals include allowing existing allocations to be reclaimed or traded to supplement the supply of IPv4 addresses to meet continuing demand. This paper will provide an appropriate contextual framing, suitable for a multidisciplinary audience, of the ongoing discussion of how market forces might be introduced to enhance the management of the Internet address space. We discuss some of the proposals currently under consideration and highlight the key points of contention. While we conclude that market-based incentives can play an important role, how they are implemented needs to be carefully considered in light of the requirements for Internet routing, the politics of global Internet governance, and the likely implications on the transition to IPv6. For example, we believe inappropriately allowing unrestricted trading of IPv4 addresses could have the perverse effect of increasing the aggregate costs of Internet routing, retarding the migration to IPv6, and thereby adversely impacting Internet growth and architecture. On the other hand, while the existing framework for Internet address management has served reasonably well to date, retaining the status quo is likely to result in a burgeoning black/grey market in IPv4 addresses and similar unfavorable outcomes with respect to aggregate costs and Internet architecture. Further research and collective discussion is needed to identify how best to forge a middle road toward improved Internet address management. The discussion provided herein is intended to help frame and seed the debate.