The end of the Internet?

By Alan Way On January 26, 2011
No tags assigned.

Were the Mayans correct in predicting the end of the world in 2012? Or did they get some mixed signals on the state of the Internet and the availability of IPv4 addresses?

For many years the pundits have been preaching imminent disaster as we near the exhaustion of the IPv4 address pool, but somehow we kept dodging the bullet. This year however there is no doubt, the end is nigh.

The Internet Assigned Numbers Authority (IANA) is down to less than 2% of the available IPv4 address pool (at the time of writing according to Latif Ladid, president of the IPv6 Forum, the IANA ran out of addresses the week of Jan 17th). It is true that the Regional Internet Registries (RIRs) may still have enough to keep them going until 2012, but it is going to get increasingly difficult to get large contiguous blocks of addresses. It has also been reported that the RIRs could run out as early as November 2011! At the last Reseaux IP Europeens Network Coordination Center (RIPE NCC) meeting held in November of 2010, they announced that this could well be the last meeting before they run out of free IPv4 addresses.

So what are we going to do? We could continue with the use of Network Address Translation (NAT), but this is not a very elegant solution even though it has been used for many years. Basically it interferes with the data between the source and destination. It can cause problems for IPSec and make it more difficult for lawful intercept, as well as creating speed bumps for the Internet superhighway, degrading performance and QoE (Quality of Experience) for the users.

IDC reported in 2009 that they expect more then 15,000,000,000—yes, that is 15 BILLION—devices to be connected to the Internet. A new term used to describe this phenomenon is the “embedded Internet.” In other words, this growth of connected devices will not be driven so much by computers, but by the new generation of cell phones that use native IP. New consumer and entertainment devices, such as tablets and Internet enabled TVs, will also help drive this new connectivity, along with more industrial and medical automation.

Network Address Translation (NAT) or Carrier Grade Nat (CGN) is not a real solution; it is only a band aide. Ultimately, IPv6 will surpass IPv4, even though this will take many years. In the meantime, the network will need to evolve. New equipment should be deployed with IPv6, while current systems need to be updated to dual stack. With dual stack, we can maintain the ability to connect with IPv4 and still connect to new devices being deployed with IPv6.

Techniques such as Dual Stack Lite (DS-Lite), although not perfect, are for now possibly our best interim hope. DS-Lite enables an ISP to provide an IPv6 service, this allows for the transport of IPv4 over IPv6 tunnels. I have not seen any large scale deployment, and there may be some large scaling issues, but in my opinion, this could be one of the best comprises for now.


comments powered by Disqus
× Spirent.com uses cookies to enhance and streamline your experience. By continuing to browse our site, you are agreeing to the use of cookies.