DoD Internet Protocol Version 6 Transformation
by Capt Michael R. Cirillo
‘Do you see the L?’
‘Yes, we see the L,’ came the response.
We typed the O, and we asked, ‘Do you see the O?’
‘Yes, we see the O.’
Then we typed the G, and the system crashed (and the Internet was born).
—UCLA, 1969
The Internet protocol (IP) gives us potentia ex indicium—power through information—and how we enable this power through IP is undergoing an exponential transformation. What is technically arguable for this transformation is not so much that the available IP Version 4 (IPv4) addresses will run out or that we should be overly concerned with when the addresses run out, but that we are too close to potentially exhausting the remaining number of IPv4 addresses. This risk, combined with future requirements, compels the Department of Defense (DoD) to begin this transformation.
The Prelude
Per the 9 June 2003 DoD memo, Internet Protocol Version 6 (IPv6), the DoD must “complete the transformation to IPv6 for all inter and intra networking across the DoD by FiscalYear 2008.” Because IP is critical to how the Marine Corps functions, this article gives consideration to some potential Marine Corps transformation concerns and addresses why we’re transforming from IPv4 to IPv6. Data communications hinges on IP, but while we don’t need to know how a car’s engine works in order to drive it, we do need to know that the engine is under the hood, has oil, antifreeze, etc. and can be counted on to run effectively. If we refer to our car as IP and, when in our car we are the information IP carries from point A to point B, we can understand IP’s criticality.
Although IPv4 still has billions more available addresses, exhausting this resource is probable. Including the entire 1990’s boom, draconian assignment policies, advancing use, and compression technologies, the best-case scenario is that IPv4 addresses are available for nearly a decade past DoD’s 2008 mandate. At worst, we have less than 5 years before IPv4 addresses run out, but Occam’s Razor1 tells us that the truth likely falls in the middle. Regardless, change is upon us.
Although some stand-alone computers exist, most connect to a network, so those who doubt IP’s criticality need only turn off their computers’ network connections forever. Computers are ubiquitous and, via the connection or networking of these computers, Marines generally understand the tactical and strategic significance of networked information. Marines know that computers affect virtually every aspect of our lives, including getting us our pay; maintaining our personal, professional, and training records; accessing Marine OnLine; MarineNet training; World Wide Web access; and of course, e-mail and chat. We hear IP’s effect on us in our jargon when we speak of and use e-mail, local area networks, applications, the global address list, routers, firewalls, wireless, Blackberries, laptops, Google, etc.
The DoD global information grid (GIG) centralizes data gleaned from many network devices and distributes this information across networks predominantly using IP (again, the car analogy). Information gained this way, via IP, transforms the warfighter’s situational awareness into shared awareness, increases decentralized decisionmaking, and reduces Clausewitz’s fog of war. Network-centric warfighters know that relevant information is the ultimate weapon because it reduces the time it takes to know, to react, and to destroy. IP makes the information weapon work better than it could ever have been conceived of working before IP. IP reduces the OODA (observation, orientation, decision, action) loop to an analyze-commit bullet.
Why We’re Transforming
There’s a DoD mandate and, for the Marine Corps, this reason is sufficient. Despite manifested tangible requirements, the DoD states that IPv6 is designed to overcome IPv4 limitations by “expanding available IP address space, improving end-to-end security, facilitating mobile communications, enhancing quality of service and easing system management burdens.” This statement’s factual foundation is based on a less than stunning revelation of the transformation’s factual advantages, so despite the vision, the devil’s in the details. Although we are not yet resting on a year 2000 (Y2K) precipice, and some IPv6 opinions do read like rampant optimism, we can be said to be resting on the top of a Yv4 slope—or the year IPv4 becomes untenable.
Although IPv4 has been in use for more than 20 years, the DoD reasons that IPv6 better supports future combat systems, heightened network ubiquity, mobility, ad-hoc networking (dynamic addressing), and security. Conversely, the DoD states that IPv4 cannot support future required capabilities and is a barrier to IPv6 implementation, but still recognizes the IPv4 sunk costs and a presumed satisfaction with IPv4. The DoD runs the largest IP network in the world and states that the “achievement of net-centric operations and warfare, depends on effective implementation of IPv6 in concert with other aspects of the GIG architecture.” However, unlike the web browser for IPv4, there is no IPv6 “killer application” demonstrating the transformation’s benefit, so vice simply playing with the technology and proving that a network-centric inflated tire is better than a solid one, again, the benefits need real exposition. Absent obvious transformation requirements or a World Wide Web blossoming, our IPv4 capability to operate and defend networks continues to serve us. Perhaps into the transformation, as the DoD and commercial sector symbiotically advance, such a killer application will appear.
Potential Transformation Concerns
The seminal transformation issue is that IPv6 is not compatible with IPv4. To use the new protocol, every networked device must change its software protocol. The DoD will make these changes through device attrition and upgrades, but they constitute a mandatory, massive and, potentially, lengthy forklift of gear. Because of the vast IPv4/IPv6 size difference, any device running IPv6, much less both protocols, requires more processing power and memory. IPv6/IPv4 translation topology changes could, in the short run, impact network speeds, slow applications, cause new software problems, and create new training requirements, all of which translate into Marine Corps fiscal, logistical, personnel, and expeditionary bandwidth issues.
A battlefield-to-the-beltway focus should be the transformation timetable’s prime consideration because service failures or interruptions can seriously impact operating, deploying, and returning units plugging in and out of transforming environments, including in-theater IPv6 use, global troubleshooting, and different scales of geographic transformation participation. A bottom-up warfighter focus ensures that tactical course corrections remain possible until every device and application is IPv6 enabled. A date prior to the DoD’s 2008 mandate should be established at which point all network devices are IPv6 compatible, and conversely, within the DoD there exists no IPv4-only devices on the GIG. A lengthy transformation risks simultaneous or domino-type massive failures of dual IPv4/IPv6 stacked devices resulting from faulty applications, operating system manufacturers, malicious intrusions, timing, latency, connectivity, and quality of service issues. Absent a continuity of operations, failing transformation efforts could deleteriously impact connectivity for operational units or remote sites. A lengthy IPv4/IPv6 translation gives rise to an IP management reality where there could be a combined IPv4/IPv6 Internet for an extended period of time.
A swifter transformation creates inertia to hasten IPv6 conversion and should impel remaining IPv4 domains, applications, devices, etc. to transform sooner. Since tactical bandwidth does not approach garrison/commercial speeds and, at the lowest tactical level, bandwidth is essentially a dribble in comparison, spending more time suffering through transition lowers combat information power. Also, since adding additional IP size and number of IP packets to existing bandwidth use may dramatically increase global network bandwidth use and the size of routing tables, this transformation bubble needs to be short-lived.
Conclusion
IP affects Marines every day, and we have to squarely address valid transformation concerns. The DoD’s transformation to IPv6 is based more on a mandate to evolve than upon what has been proven correct or incorrect, and there are considerable strategic network topological changes that will and will not occur resulting from what IPv6 does and does not do. The evolutionary scope of this transformation connotes an ever-expanding reliance on network-centric warfighting in which there is considerable room for what is undecidable, unknowable, and theoretical.
In our networks, applications, and systems we should begin seriously gaining experience with IPv6. This piloting and testing puts IPv6 use in a networking crucible and gives the Marine Corps the ground truth on IPv6 benefits and shortcomings. This transformation is potentially greater than anything the DoD has previously contemplated, so proven, substantial reasoning must exist prior to charging down the road of technological change or, worse, procrastinating until migration goes from a Yv4 slope to a Y2K-type issue.
Note
1. Occam’s razor is a principle attributed to 14th century English logician and Franciscan friar William of Ockham and states that the explanation of any phenomenon should make as few assumptions as possible, eliminating those that make no difference in the observable predictions of the explanatory hypothesis or theory. Definition from <www.wikipedia.org>.
>Capt Cirillo is the Network Operations Officer and Information Assurance Manager, III MEF.
