As covered in 4G Digest’s October 13th issue, IPv4 (Internet Protocol version 4) addresses – the basis for Internet as we know it today – are quickly running out. During a historical event celebrated last Thursday in Montevideo, Uruguay, the last five blocks of vacant IPv4 addresses were equally allocated to the five Regional Internet Registries (RIR), which manage Internet addressing in the five regions in which the world is divided. Now each RIR will be responsible for carefully assigning its last block among content providers, Internet access providers and other companies. The APNIC (Asia Pacific Network Information Centre) forecasted it would lead the IPv4 resource exhaustion because of its accelerated demand for IP addresses.

Although there is consensus on the need to migrate to IPv6, the consequences of the long-announced end of IPv4 are usually not clear. As stated during the Montevideo conference, the Internet will not suddenly stop working when the very last IPv4 address has been assigned. The situation was compared to what an exhaustion of driving plates would mean for road traffic: we would not notice in the short term, but many end-users would complain in the long term. The message from conference host ICANN (Internet Corporation for Assigned Names and Numbers) was the migration from IPv4 to IPv6 should be completely transparent to end-users, although the real implications for them during the transition period would depend upon the compromise from content providers to offer unified services over IPv4 and IPv6, and from Internet access providers to guarantee proper communication for both IPv4 and IPv6 customers.

So if both IPv4 and IPv6 users will be able to co-exist, why should existing IPv4 allocations be ever migrated to IPv6? Some light on top of this question was shed in Montevideo by the most respectable minds in the Internet world. The main risk when not migrating to IPv6 is the loss of the end-to-end communication that today’s homogeneous Internet allows. If alternative address translation methods between the two protocol versions are broadly adopted, instead of directly replacing IPv4 by IPv6, our daily communications (especially real time services like VoIP and video) could be disrupted. Furthermore, users remaining with the old IPv4 could feel discriminated against because of a lower quality Internet access caused by the intermediate protocol translations each data packet would have to go through. As any other network technology, early adoption of IPv6 will allow content providers and access providers to differentiate from other companies delaying the transition.

This differentiated network performance among operators adds to today’s trend of increasing networks’ flatness and transparency. LTE, the last word in mobile networks, ensures a flatter topology allowing for reduced latencies that should enable new applications such as on-line gaming. Network neutrality is also related to this, being one of the industry’s hottest topics. Multinational groups Vodafone and Telefónica are continuously claiming their intention to charge service providers for traversing their pipes. In North America, Comcast won the trial against the FCC because of P2P traffic blocking and Verizon’s and MetroPCS’s recently filed lawsuits in the same direction. The need to use traffic policing techniques is not an option for operators, since their own economical survival is threatened by the so-plotted diverging revenue and cost curves. Furthermore, a delayed adoption of IPv6 will have a direct impact, since the additional equipment needed to do protocol conversion will increase costs in the long term and a lower performance network could reduce revenues by stimulating churn. IPv6 may not cause a new Y2K Problem, but will pose an additional challenge for an industry that is still agreeing on the prescription for current data indigestion.

MARAVEDIS is a leading analyst firm focusing on 4G and broadband wireless technologies and markets.

Author: By Esteban Monturus, Market Analyst - Europe & Backhaul