The impact of the dwindling supply of IPv4 addresses will depend on how long it takes to deploy IPv6 fully. It is in that interim period that we'll see whether cloud computing can really fulfil its promise of shared resources, says Lori MacVittie.
The last blocks of IPv4 addresses were distributed in a widely viewed ceremony of sorts. While the distribution of these blocks to regional registries does not mean that there are no more IPv4 addresses available, it does point to a time when all addresses will be depleted. What does the prospect of the depletion of IPv4 addresses mean for organisations?
Not much for those who already have blocks of addresses reserved for their use. But for start-ups or organisations that have grown large enough to require their own presence, it may mean that they will be forced to look to hosting providers and cloud computing until IPv6 deployment is complete and the infrastructure of the internet is prepared to support them.
IP address supply and demand
It is expected that the last blocks of IPv4 addresses will actually be depleted in the next six to nine months. As available addresses dwindle, it is naturally the case that the cost of obtaining even a small block for organisational use will increase. And when they're gone, they're gone — unless generous individuals who were assigned large blocks of addresses when they were cheap and freely available continue to give back some of the addresses for distribution.
Ultimately this situation means the supply of IPv4 addresses available for public addressing of public-facing services — sites, applications and DNS — is rapidly dwindling. Until IPv6 is more fully deployed and a healthy influx of addresses becomes available, it bodes ill for organisations currently without their own public space on the internet.
For some — especially start-ups and smaller businesses experiencing growth — this shortage could mean trouble. If these organisations haven't already set aside some address space for future use, they may need to turn to hosting or cloud-computing providers for their public-facing services.
Cloud computing and shared resources
When cloud computing is the topic of discussion, the phrase "shared resources" generally brings to mind compute and storage. Network, too — but more in the sense that organisations deploying applications in a cloud-computing environment must share physical bandwidth.
We rarely discuss IP addresses as shared resources but apart from specific offerings, that is exactly what happens in the cloud. IP addresses are shared unless you specifically request otherwise. As it becomes more expensive and difficult to obtain a block of IPv4 addresses from local providers, organisations needing additional or new blocks may be forced into the cloud, whether they wanted to be there or not.
That's because IP addresses in large blocks are available to cloud-computing and hosting providers and can be more easily leased to their customers to enable public-facing services. Thus, organisations attempting to start up their public presence just when there may be a dearth of IP address availability — between IPv4 depletion and IPv6 deployment — may find...
...themselves offered few options other than outsourcing their public presence to obtain a public-facing presence.
The question is whether they will ever go back once IPv6 deployment is complete and addresses are available, or will those who do move to the cloud because they need a public-facing presence stay there, having invested in systems and architectures specific to addressing the challenges of a shared-resource environment?
Growth in service-layer virtualisation
For organisations that already have public IPv4 addresses, we will probably see an increase in the use of virtualisation — the same technology exploited by cloud and hosting providers to share those resources. Not the server or operating-system virtualisation that comes to mind most often when the technology is mentioned, but the service-layer virtualisation that has existed since the early days of network-address translation and load-balancing as a means to achieve scalability.
Most large organisations already employ service virtualisation — the practice of enabling multiple virtual-service hosts on the same IP address and port combination — and are using network-hosted technology such as load-balancers and application-delivery controllers to steer the traffic to the appropriate service host.
The impact of the dwindling supply of IPv4 addresses will be directly related to the length of time between depletion and full deployment of IPv6. It is in that interim period that we will see whether cloud computing can really follow through on the touted benefits of shared resources.
There is unlikely to be again a situation in which such a valuable resource will be in such short supply. Whether cloud computing or hosting providers will benefit depends largely on whether the perceived risks of cloud computing outweigh the benefits.
Of course, the rest of the industry might shift into high gear to deploy IPv6 and avoid the problem altogether. But given the scale of that switch, it is a possibility that unfortunately seems a lot less likely.
Lori MacVittie is responsible for application services education and evangelism at application delivery firm F5 Networks. Her role includes producing technical materials and participating in community-based forums and industry standards organisations. MacVittie has extensive programming experience as an application architect, as well as in network and systems development and administration.