Convergence applications bear some of the direct and indirect responsibility for the massive increase in demand for Internet addresses. The situation is getting critical, and one of the doyens of the Internet -- Vint Cert -- is worried.
Currently, the Internet uses an addressing scheme called Internet Protocol version 4 (IPv4). For years, experts have been treating address exhaustion like the IT version of global warming. ("Things seem fine now. But once the crisis hits, you'll all be sorry that you didn't listen.") Adoption of the next addressing scheme, IPv6, has been slow.
Cerf, who is the outgoing chairman of the Internet Corporation for Assigned Names and Numbers (ICANN) and the chief technology evangelist for Google, says that the address cupboard will be bare by 2010 or 2011. He was somewhat reassuring, however: Exhaustion of the available supply doesn't mean that the Internet will grind to a halt. It simply means that new addresses won't be available.
The answer is IPv6, which will make available a trillion trillion trillion discreet addresses (IPv4 offers a paltry 4 billion). Ars Technica both shines light on a related issue and highlights how complex the overall field is. The story says that organizations controlling address distribution in North America and Europe, The American Registry for Internet Numbers (ARIN) and the Rseaux IP Europens (RIPE), respectively, have come out in favor of a move to IPv6.
That's the easy part. The complexity of the addressing sector is illustrated by the fate of the class E space addresses in IPv4. The story says that this block of licenses -- which start with the numbers 240 to 255 -- were reserved for future use. Combining these with the still available standard IPv4 addresses, the story says, could buy the older standard an extra year.
The key takeaway isn't what happens to the E space addresses. The key is the realization that anything involving a change as basic and labor-intensive as tweaking millions of routers and other devices is prone to get a wee bit contentious.
This level of uncertainty isn't likely to be resolved soon. For instance, Johna Till Johnson goes a long way to deconstructing the common wisdom that we are running out of IPv4 addresses. She adds that there is a price in switching to IPv6, since the longer addresses (necessary to get from 4 billion to multi-trillions of possibilities) requires more bandwidth and could be a problem in lower capacity connectivity scenarios.
In cases in which there is uncertainty over whether or not to deploy a technology, it is a good idea to see what vendors are doing. These are business folks who certainly won't invest in something that they don't feel will happen. Last week, Agilent introduced N2X multiservices test gear. The company says that it is the first product geared to validating Internet protocol television (IPTV) over IPv6 networks. The release provides a tremendous amount of information on the attributes of the N2X. The bottom line is that a major test and measurement equipment vendor considers such an offering worthwhile.
EMC Corp. also is at least hedging its IPv6 bets. In September, the company unveiled the Smart IPv6 Availability Manager. The software, EMC says, offers network discovery, mapping, monitoring and root-cause analysis for IPv6 as well as IPv4 and hybred infrastructures. The product meets U.S. government requirements for IPv6 support and has been submitted for Ready v6 Logo testing. The release provides details about the product, which is positioned as a way to achieve visibility and control in complex IP environments.
IPv6 seems to be a matter of when, not if. The fundamentals are clear: There is an insatiable appetite for slapping an IP address on just about everything. The challenge is getting network operators to do something that is expensive, troublesome and revenue-neutral before it becomes a crisis. Calls like those from an expert such as Vint Cerf certainly will help. Whether it will be enough to finally get IPv6 over the hump remains top be seen.