The IPv6 Business Case
This article discusses what IPv6 means for your business. Should you invest in IPv6? If yes, when is the time to do it? How can you plan for IPv6 in order to make your transition a smooth and cost-effective one?
Obviously, when you introduce IPv6 into a network, costs will initially rise. You have to educate your IT staff on IPv6, you have to build test beds that let you test IPv6-related issues, you have to create a strategy, and you also have the costs of implementation.
And what is your return on investment? Why should you invest in IPv6 while you have a running IPv4 network?
A recent study conducted by the Department of Commerce (1) in conjunction with the independent research group RTI International (2) examines the costs and benefits associated with the transition from IPv4 to IPv6. The study comes to the following conclusions:
The cost for transition to IPv6 is estimated to be $1 billion per year, so IPv6 creates a services market projected to be approximately $25 billion over the next 25 years. The financial benefits of IPv6 transition are expected to be $10 billion per year. Of each dollar invested in IPv6, only about 8% is projected for the actual infrastructure upgrade and the other 92% for leveraging the advantages of IPv6.
The cost savings are expected in several main areas:
The costs for network hardware and software upgrades are considered to be negligible. Over the next four to five years, most network hardware, operating systems, and IP-enabled software will support IPv6. These upgrades can therefore be done as part of routine upgrade cycles. In fact, many of the products already do this. Most router vendors have implemented IPv6 since 1998, and have continuously enhanced and optimized their implementations. Also, most operating systems have supported IPv6 for quite a few years. Microsoft has committed to support IPv6 for a long time.
Windows XP with SP2 includes an IPv6 production stack with some limitations, due to the networking architecture of XP. In Windows Vista IPv6 is enabled by default and configured to be the preferred protocol. The limitations of Windows XP have been removed. Windows 2003 Server supports IPv6 and Longhorn will include a production IPv6 stack. Apple's OS X has supported IPv6 since 2004, and so have most of the Linux and Unix flavors.
By January 2006, the total number of IPv4 addresses still available at IANA and the RIR is 1,300.65 million addresses (3). In 2006, 167.96 million addresses have been allocated. If we project this allocation rate into the future, we run out of IPv4 address in approximately seven to eight years if there is no change in allocation policy and address demand. With the mobile services evolving, we can expect a major increase in address demand in the near future. One area of increased address demand will probably be IP telephony on a global basis, and the development of IP-based monitoring and control devices across many different industries.
Another area of increasing address demand is the development in Asia. While people in the developed countries think that there is no life without the Internet, actually only about 15% of the total world population is connected to the Internet (4). Asia for example, where more than 50% of the world population lives, has an Internet penetration rate of 10%, where, by comparison, the Internet penetration rate in the U.S. is 69%. If we want to connect only 20% of the world population to the Internet, we would need some 390 /8 (Class A) address blocks. This is far more than currently available.
By January 2006, 50,809 /32 blocks of IPv6 have been allocated (3). Note that each /32 block is larger than the whole IPv4 address space. With this allocation, 3.3 billion customers can get a /48 address block. However, this represents less than 0.01% of the available IPv6 address space. So the saying that IPv6 has multiple addresses per grain of sand on the planet is really true - this is the way we will be able to meet the address demand coming up in the near future.
What should you do today?
There are many different ways to integrate IPv6 depending on the infrastructure and the requirements. It is expected that a large part of our networks will be dual-stacked for many years. This means that we will have to find ways to master the parallel management of two protocols. Major areas for careful planning are address management and security.
Today, you do not lose connectivity in the global Internet if you don't have IPv6. But this will change in the near future. So while there is time, I advise you to do your education and your research and create a good transition plan for your network, to make sure you are ready when the time comes.
Silvia Hagen, Owner and CEO of Sunny Connection AG, is the author of a number of books, including, "IPv6 Essentials," published by O'Reilly. She regularly speaks at international technical conferences in the United States and Europe.
Sunny Connection AG (www.sunny.ch) is a leading IT consulting and education company based in Zurich, Switzerland. Its main expertise is in directory services integration, Identity Management and in network and protocol analysis, with more than 10 years of experience in consulting to mid-sized and large companies, mainly in the areas of industry, banking and insurance.
If you would like to discuss specific issues with us or to have an in-house, customized education for your IT decision makers, strategists and IT engineers, please contact us at www.sunny.ch or email to firstname.lastname@example.org.
1) Technical and Economic Assessment of Internet Protocol Version 6,
Department of Commerce, February 2006, http://www.ntia.doc.gov/ntiahome/ntiageneral/ipv6