First impression on unpacking the Q702 test unit was the solid feel and clean, minimalist styling.
IPv6 migration: Do it for the right reasons
- — 07 September, 2011 00:46
This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter's approach.
Much of the conversation around IPv6 has been based on the fear of IPv4 address exhaustion and the impending collapse of the Internet if we don't migrate. If we don't comply, customers will be unable to reach our sites and we will simply disappear from the electronic world.
Fear, however, is a poor motivator; it does not build a business case.
Most practitioners agree that the word "migration" is a misnomer in conjunction with IPv6. It suggests moving to a pure IPv6 world and eschewing the existing IPv4 world. Experts would agree that this simply isn't true, practical or even possible for some time.
IPv4 is the standard, and even if the world at large is actively moving to IPv6 as a standard, many of the most highly desired content is still in the IPv4 world. Chances are IPv4 will be around for quite some time.
So the real question is: "What does moving to IPv6 mean for my bottom line?" How can moving to be IPv6 be a positive business investment instead of simply a cost of doing business?
First things first
The business value of any endeavor has two components: the benefit and the cost. Even if the benefit is significant, if the cost is greater, then it is difficult to make the justification. However, even if the benefit is small and the cost is insignificant, it just might make sense.
So, what is the real benefit of moving to IPv6? In general terms, the business case simply doesn't exist. According to some research, IPv6 traffic over the last year has accounted for less than 0.3% of all the Internet traffic regardless of the source.
It is important when examining this that you understand the current IPv6 market. Very few Internet Service Providers (ISPs) currently deploy IPv6 to the consumer market, meaning that IPv6 is not a necessary mechanism to reach most general consumers.
This is equally true of hosting companies where many organizations have found it difficult to even get connectivity to the IPv6 Internet and have had to find alternative means to participate.
Even those service providers (outside NA mostly) that provide direct IPv6 addressing to consumers must still provide access to the existing IPv4 Internet and simply NAT the addresses to IPv4 -- often with no connection to the IPv6 Internet at all. B2B is likewise similar. Most organizations, even if they internally support IPv6, are not directly connected to the IPv6 Internet as the standard is still IPv4.
Mobile is often considered the primary driver of IPv6. Many mobile service providers have had to transition their mobile networks to IPv6 due to the massive growth of mobile devices, the limited number IPv4 addresses they can use and the need/desire to maintain IP addresses as devices move around the network. In addition, many of these mobile devices also have a natural preference for IPv6 over Wi-Fi networks if available.
Once again, however, since the bulk of content being consumed on these devices is still served via the IPv4 Internet, most mobile providers also do not directly connect their IPv6 client networks to the IPv6 Internet.
This situation, above all others, is the reason behind the slow adoption of IPv6. The business case behind providing support for IPv6 is either based on investment in the future with possible "first-mover" advantage in providing IPv6 content/services, or in specific applications where providing IPv6 connectivity gives your offering an advantage.
The former case is difficult to quantify and certainly does not suggest a complete "migration" to IPv6 as it may be some time before consumers can even reach the IPv6 service. The latter case is much easier to quantify depending on the business need.
For example, if you provide content services for mobile devices, enabling native IPv6 connectivity directly to the mobile providers (not over the IPv6 Internet) can give you significant advantage in performance/reliability while making it easier for those providers to manage their networks; this IPv6-ready strategy will eventually also pay off for IPv6 Internet connectivity.
What about the cost?
Since many business organizations are fixated on the lack of a well defined business opportunity, they fail to look at the other side of the equation: the cost of providing IPv6 services, especially the cost of doing it today rather than tomorrow or somewhere down the road.
Because the conversation about the eventual requirement to transition to IPv6 has gone on for more than a decade (the IPv6 standard was published in 1998), most infrastructure equipment is generally capable of running IPv6 today.
Not only that, but many organizations already have devices in their network that can easily provide dual-stack (IPv4 and IPv6 simultaneous connectivity) solutions or IPv4 to IPv6 translation. Even the venerable DNS software BIND has fully supported IPv6 for over a decade and supports DNS64 for dual-stack, transitional solutions.
In reality, the cost of adopting IPv6, at least from a capital expenditure point, will often be quite negligible for most organizations as it is already integrated into their existing equipment; from switches and routers to firewalls and application delivery controllers alike.
COST ANALYSIS: Getting at the real truth about IPv6
There was a time when IPv4 address space was free and even a longer time when it was inexpensive. As the address space became scarcer, the cost of acquiring it increased to the point that Microsoft reportedly paid $7.5M for a block of IP addresses from the defunct Nortel Networks in March of this year.
While scarcity is generally not something we are concerned about with IPv6 addressing (there are dozens of attempts to quantify the number of IPv6 addresses, but most of them come down to "virtually inexhaustible"), now that we have become accustomed to paying for them it is unlikely they will remain free forever. In fact, there is already discussion about "vanity IPv6 addresses" -- where there's a buck to be made, someone will find a way to make it. Regardless, you can rest assured that getting into the game sooner rather than later will be more cost effective.
Now ... Why?
Why not? Every day there will be more and more services available over IPv6 and more users accessing it natively. The transition to IPv6 is happening and the adoption rate is likely to accelerate as it reaches an inflection point and the power of the IPv4 Internet begins to wane.
When is the "right" time to get involved? You can certainly wait until you can make a more defined business case, but the costs will also be much greater, there will be a time constraint, and your competitors may very well beat you there.
If, on the other hand, you begin strategically making the move today, you have time to figure things out. When you get involved at the ground floor the costs are guaranteed to be minimal compared to the unknowns of the future. So, while the business case remains uncertain, the cost for many organizations is also minimal; the time to ROI may be unknown, but its inevitability is guaranteed.
Why put off until tomorrow that which you can do today, especially when it is cheaper and you know it has to be done?
Read more about lan and wan in Network World's LAN & WAN section.