The IPv6 movement has been years in the making. So many years, in fact, that it has hardly been a movement at all. While a handful of regions, primarily in Asia and pockets of Europe, have embraced IPv6, it has been otherwise largely ignored, something to be considered later while we exhaust IPv4 assets. This thinking has clearly stunted the growth of IPv6, presenting opportunities to early adopters and IPv6 facilitators and indigestion for the procrastinators.
Besides the indepletable address space (approximately 340 trillion, trillion, trillion unique IP addresses vs. IPv4's 4.3 billion), IPv6 offers a number of network advantages: In most cases, for example, computers and applications will detect and take advantage of IPv6-enabled networks and services, and in most cases, without requiring any action from the user. And IPv6 relieves the need for Network Address Translation, a service that allows multiple clients to share a single IP address, but does not work well for some applications.
For the Internet to take advantage of IPv6 most hosts on the Internet, as well as the networks connecting them, will need to deploy the protocol. However, IPv6 deployment is proving a bigger challenge than expected mostly due to lack of interest from the service providers and end users.
While IPv6 deployment is accelerating, according to a Google study, areas such as the Americas and Africa are comparatively lagging in deployment. In December 2010, despite marking its 12th anniversary as a Standards Track protocol, IPv6 was only in its infancy in terms of general worldwide deployment.
There are indeed interoperability issues between IPv6 and IPv4 which is leading to the creation, essentially of parallel, independent networks. Exchanging traffic between the two networks requires special translator gateways, but modern computer operating systems implement dual-protocol software for transparent access to both networks either natively, or using a tunneling protocol such as 6to4, 6in4, or Teredo.
Even early adopters of IPv6 as a network service understand that demand doesn't exist across the board today, but will soon. Akhil Verma, director of global product management for Inteliquent (formerly Neutral Tandem/Tinet), says, "IPv6 service providers lack the content, infrastructure and applications to make a good business case today and have contributed significantly to the lack of IPv6 adoption." This coming from Inteliquent, one of the world's leading providers and facilitators of IPv6 network-enabled solutions. Once the inevitable IPv6 levy breaks, an overwhelming number of network and content providers will be scrambling (and likely overpaying) to get IPv6 equipped and compliant -- a potential boon for those few IPv6 veteran IPv6 enablers.
"We have been doing IPv6 for over a decade now, and have been enabling it for our customers on demand as a complimentary component of the services we offer," Verma says. "In turn, this means any IP Transit customer can get IPv6 service enabled on existing ports so they can have dual stack access at any time. We continue to enable services for new IP transit customers, but we see a very low demand. It could be an education issue, financial justification issue or some other factor. Whatever it is, it is stopping customers from making the move."
According to Dr. Kate Lance, communications manager, IPv6Now, an Australian based company specializing in IPv6 assistance, training and consulting, in future years we will look back on this IPv6 debate with astonishment and ask, How could anyone conceivably argue against a technology that offers secure communication to people and devices on a breathtaking scale? How could anyone not want a technology that offers an incremental leap in Internet capability and capacity that may be as significant as the development of the Internet itself?
But challenges related to IPv6 implementation remain, relating specifically to transition complexities, costs, timelines and the overall business of the migration from IPv4 to IPv6. These challenges include:
* Operational challenges: The operational challenges are, actually, no different from the normal challenges of running any network: Staff training and equipment updates take place as in the normal business cycle. All modern network equipment is IPv6-ready -- it just needs the staff to run it. The actual transition -- like any technical upgrade -- takes planning and resources, but in operation it quickly becomes a standard operational environment. Still, for those without the necessary recourses or in-house expertise, IPv6 implementation challenges can be overwhelming ... and expensive.
* Transition and implementation challenges: IPv6 has been hamstrung by bad press, says Lance, especially the view that IPv6 is a purely technical transition without financial justification. Communication has been a disaster between levels of organizations: Businesspeople, who comprehend the strategic implications, have not been in effective communication with network people, caught up in the technical details, who fail to understand what IPv6 can offer business at a higher level.
Once an organization decides it will move to IPv6, the greatest transition hurdle is overcome! All reports so far are that IPv6 is easier to implement than people fear, and the benefits of efficiency and economy quickly become apparent. It's not an all-or-nothing environment: IPv6 can be implemented in stages as required, Lance says.
* Financial investment justification: IPv6 networks are easier to manage than IPv4 networks, so ultimately less expensive. For instance, merging two networks when businesses combine or expand becomes extremely easy under IPv6, but under IPv4 today can be costly. This is because most internal IPv4 networks use the same RFC1918 private number space, so a merger means an expensive renumbering exercise to avoid clashes and black holes.
In IPv6 the vast address range means this would never happen. In fact, any network reorganization or expansion becomes easier due to IPv6's numbering scheme, which over time means cheaper networks. IPv6 also has mobility and security features that when implemented have enormous implications for financially beneficial innovation.
* Internal educational challenges: It's important for staff who interact at a technical level with customers to at least understand the existence of IPv6 and some of the issues, as more and more customers will start using it. Managers of call centers, and managers of software and hardware groups will need some education. But the deepest level of knowledge required is at the system/network level, and staff there need some formal training.
* Availability of knowledgeable workforce: Many technical people are aware of the need for IPv6, but have been held back by other demands and business priorities. The knowledgeable workforce is not large at the moment, but IPv6Now's experience from numerous training courses is that people find it easier to understand than originally expected, and quickly develop confidence and expertise. This does, however, mean either hiring a knowledgeable staff or training-up existing staff.
According to multiple sources with firsthand experience regarding both IPv6 network transition and equipment challenges, a seamless IPv6 transition will require a systemized approach that leverages deep telecom and IP network knowledge across multi-vendors and technologies. This systematic approach is critical to developing an efficient, comprehensive and hassle-free IPv6 implementation plan, design and execution. Moreover, it is key to facilitating a smooth IPv6 migration strategy across diverse architectures.
For most, this means partnering and collaborating with an established IPv6 vendor that can bring valuable implementation insights and provide guidance on avoiding any operational challenges that will plague the majority of the "go it alone" IPv6 implementers.
There are a number of different paths one can take when making the transition from IPv4 to IPv6, each having its own set of merits and/or implications for service providers. Simplest and easy, the most logical option for almost all service providers is to deploy dual-stack on their network allowing support for both IPv4 and IPv6 simultaneously. This can be deployed in different ways, depending on the service provider's vision and network capability.
The frequently used methods are:
* Option 1: Dual stack throughout the network: The most commonly used option, whereby the network is designed to support both native IPv4 and native IPv6 simultaneously to the end customer. The service provider can offer IP transit service in multiple flavors namely, native IPv4, Dual-Stack IPv4/IPv6 and native IPv6. The native IPv4 is business as usual practice being run for ages, Dual-Stack IPv4/IPv6 allows the service provider to offer both IPv4 and IPv6 services over the same backbone where both kind of traffic simply transverse the network as an IP packet. Finally, for the technology savvy end customers, the service provider can also offer native IPv6 service, which allows the end users to use the service separate to their IPv4 service, and is preferred by some customers to balance their network hierarchy.
* Option 1-B: Tunneling IPv6 over IPv4 using 6RD (anycast): While not the best way to deploy IPv6, but preferred by service providers who want a low investment and do not have an access network capable of supporting IPv6. Native IPv6 is not supported in this scenario, thus potentially requiring a second upgrade in the future, hence requiring another round of financial justification, education, etc., for the service provider. There are also other ways and means available to service providers to enable their networks to support IPv6, but those are neither easy nor common in terms of approach.
In closing, it is fair to suggest that IPv6's greatest challenges are not technical or economic, but rather the bizarre myths that have arisen, including, Lance says:
* The belief that IPv6 is just an unprofitable switchover in Internet technical plumbing, when it has the great benefit of supporting vastly larger, safer, more efficient -- and hence cheaper -- network infrastructure.
* The fear that IPv6 is a costly "all or nothing' transition" when in fact there are staging solutions to help run IPv4 and IPv6 networks in tandem.
* And the assumption that IPv6 is just like IPv4 only bigger, when in fact it needs good levels of understanding through training and workshops to be implemented securely and with maximum return on the investment.
Once a decision is taken to move forward with training and implementation then IPv6 all falls easily into place.