John Shipp, Deputy Director of Technical Architecture, Army Architecture Integration Cell, Office of the Secretary of the Army explained the IP6 transition status and planning at the Army. He said Army transition totally synchronizes with Department of Defence (DoD) effort, but complete transition of the whole network to IPv6 by 2008 was too costly.
Shipp said DoD recognizes that IPv6 is critical as the key foundation for network-centric operation and warfare, where everything gets networked.
Major future force systems, such as future combat system, joint tactical radio system, and war fighter information network tactical, will be deployed in the time frame that IPv6 starts to proliferate, so they need be built with an ability to communicate in IPv6 environment.
For DoD and Army, address space is not so important, but auto-configuration, security, quality of service, and mobility are. But they are yet to see how IPv6 can actually realize these benefits.
So far, the Army issued Initial Guidance in November 2003, with additional guidance in April 2004. Army Transition Plan is due on June 30, 2004, for resources needed in budget and POM submissions. IPv6 architecture needs to be developed by September 2004, and get resources through presidential budget by September 2004 and mini-POM (Program Objectives Memorandum) by March 2005. Actual implementation begins in October 2005.
Now, we have the architecture that defines how to put those transition mechanisms. Then well turn on IPv6 within Army Enterprise but its totally dependent upon DoD IPv6 transition. They will tell us when each system is going to transition, and we will synchronize our effort to go IPv6, said Shipp.
The Army needs to make core infrastructure dual-stack first wherever feasible, and then edge networks where technically feasible and cost effective. [While] IPv4 and IPv6 will co-exist for a long time, we try to reduce the time period as much as possible.
But transitioning all Army networks to IPv6 by 2008 is too costly, added Shipp. For the initial guidance in October 2003, the Army estimated the cost of moving to IPv6 with the assumption that transition to IPv6 in FY2008 was a mandate, and IPv6-capable commercial off-the-shelf products were available. The estimate of total cost amounted to more than 1.8 billion dollars. So they made another estimate in the additional guidance with a different assumption. This time, Transition to IPv6 by 2008 was a goal instead of mandate, with possible completion date in 2012. It was also assumed that transition would make use of technical refresh (system with 5 years of obsolescence need not upgrade), and network with lower tactical importance will not transition. The new assumptions reduced cost estimate significantly, but it could still be too high, Shipp said.
He said the Army will continue to develop mature transition plan, with effort to complete transition within existing budget if possible. On the other hand, it will reduce risks by various efforts including study the impact of IPv6 on combat systems and participation in Moonv6.
Key to implement army architecture is the guideline on transition mechanisms, when to use them, pros and cons of using each mechanisms, being developed jointly by Defence Information Support Agency, NAv6TF, Army Communication Electronics Research, Development & Evaluation Command, etc). Once we did that, we want to find the most cost-effective way to transition to IPv6. All of our networks may not need to transition to IPv6 or even be dual stack.
この記事のトラックバックURL
http://www.ipv6style.jp/trackback/384