The Quiet Revolution of Software Defined Networking at DoD

Posted by Alexander Rossino on May 23, 2016


Back in the day, a new systems architecting approach called Services-Oriented Architecture (SOA) was all the rage in the Department of Defense. You couldn’t attend a conference, read a budget document, or peruse the media without encountering comments on the benefits of SOA and the need for the DoD to move quickly toward implementing it. Then, almost as quickly as the hype cycle had ramped up, mention of SOA all but disappeared.

What happened to SOA was that it became so common it no longer drew attention. As SOA proliferated, SOA-based projects slipped into the background operating ghost-like behind the scenes. What once was hype became ordinary and ordinary doesn’t garner headlines.

The example of SOA is instructive for another technology approach that has drawn a lot of attention in recent years - Software-Defined Networking. SDN is often touted as the next-generation solution for managing and securing networks across the Defense Department. Two years ago – an eternity in technology terms, but merely the blink of an eye for a slow-moving bureaucracy like the DoD – Terry Halvorsen, the then new DoD CIO, told the 2014 Federal Forum in Washington, D.C. “We [the DoD] have to embrace … software-defined … networks – that includes all the infrastructure.” The DoD CIO was on board with SDN. Great, but what’s happened with SDN since then? Is there publicly available evidence that the DoD has embraced SDN to a greater extent or is it dissolving into the background like SOA did a few years back?

A survey of the DoD’s Fiscal Year 2017 Research, Development, Test, and Evaluation (RDT&E) budget request reveals only four programs that explicitly mention SDN as part of the technology solution being explored. These programs are spread across the Army, Navy, and the Defense Information Systems Agency.

Army – Future Communications and Networking Technologies. The requested budget for FY 2017 is $7M. FY 2017 plans include maturing “software-defined networking architecture for Army tactical edge networks and mature SDN waveforms that will identify and mitigate network vulnerabilities.” Work on this program is performed at the Army’s Communications-Electronics Research, Development, and Engineering Center (CERDEC).

Navy – Tactical Networking and Network Control/Management. The requested budget for FY 2017 is $7.1M. FY 2017 plans include “continuing the development of software-defined networking capabilities for tactical platforms.” Where this work is being performed could not be readily determined.

DISA – DoD Information Network (DoDIN) Systems Engineering and Support. The requested budget for FY 2017 is $10.3M. FY 2017 plans include advancing “next generation software-defined networks for automating and virtualizing the DODIN.” This project comes out of DISA’s Office of the Chief Technology Officer (OCTO).

Requested funding for these programs totals a mere $14M, but budget requests don’t necessarily tell the whole story. DISA has introduced SDN into the Joint Information Environment via the Global Information Grid Services Management – Operations (GSM-O) contract (#HC102812D0021) awarded to Lockheed Martin in June 2012. Similarly, the Army’s Network Enterprise Technology Command (NETCOM) will be awarding a contract in FY 2017 for the Global Enterprise Fabric program, which has software-defined elements.

These are but a few examples. The point is that while work introducing SDN into the DoD’s networks is progressing, it is often hidden inside of larger sets of requirements. This renders SDN close to SOA from a technology adoption perspective. In the long-run, SOA seems to have simply become an agreed upon standard not worth mentioning. SDN appears to be headed in the same direction; that of a quiet revolution in network management and defense of which no one takes special notice.