outcomes. Informal reports indicate that when DoD shifted funding away from university IT R&D, researchers in many areas key to DoD’s software future scaled back their research teams and redirected their focus to other fields that were less significant to DoD in the long term. The impact of R&D cutbacks generally (excluding health-related R&D) has been noted by the top officers of major IT firms that depend on a flow of innovation and talent.
This letter report summarizes three technology areas where DoD has “leading demand.” In these areas, DoD’s requirements are more sophisticated and cutting-edge than those in the rest of the marketplace. Technological advancement would significantly benefit DoD’s ability to produce the software it needs, providing a clear rationale for DoD investment in the needed research.
One area where the committee believes that new research would benefit DoD is the management of engineering risk in unprecedented large and ultra-scale systems. Such systems have engineering risks associated with early design commitments related to system functionality, non-functional attributes, and architecture. The research would focus on ways to mitigate these engineering risks at early stages of the process through new approaches to early validation, modeling, and architectural analysis.
The second area where DoD has leading demand and could benefit from technological advancement is software quality assurance for defense systems. Software assurance encompasses reliability, security, robustness, safety, and other quality-related attributes. Defense systems often include commercial off-the-shelf components and may involve global development—global sourcing is a reality for major commercial software products and, additionally, for commercial custom software and service provisioning. The needed research would focus on new ways for producers and consumers to create (and validate) a body of technical evidence to support specific claims in support of an overall judgment of fitness.
The third area, which is just as important as the first two, is the reduction of requirements-related risk in unprecedented systems without too great a sacrifice in systems capability. The challenge in this area has two parts. First, how can consequences of early commitments related to functional or nonfunctional requirements be understood at the earliest possible time during development? And, second, how can we make “requirements” more flexible over a greater portion of the system life cycle? The committee believes that the most useful research for DoD would look at ways to achieve early validation—for example, through modeling, protoptying, and simulation—and also look at how iterative development cycles can be supported more effectively and, from the standpoint of risk in program management, more safely.
These and other areas will be elaborated in greater technical depth in the committee’s final report, which will answer in more detail the questions of where management attention and research activity should be focused.