No abstract

No abstract

Glossary of Acronyms 65References 67iii | CMU/SEI-2010-TR-003 List of Tables Table 1: MDS Architectural Themes and Associated AADL Capabilities 3 List of Figures Executive SummaryThe aerospace industry is experiencing exponential growth in the size and complexity of onboard software. It is also seeing a significant increase in errors and rework of that software. All of those factors contribute to greater cost; the current development process is reaching the limit of affordability for building safe and reliable aircraft and spacecraft. The size of software in aircraft with respect to source lines of code (SLOC) has doubled every four years since the mid-1990s; the 27 million SLOC projected for 2010-2020 is estimated to cost more than $10 billion. Studies into the role of software in spacecraft accidents and the increasing complexity of flight software indicate the need for improvement in requirements elicitation and architecture, in particular for validation early and throughout the life cycle through modeling and analysis that complement testing.In order to improve predictability, the system and software engineering communities are practicing model-based engineering, where models of different aspects of a system are developed and analyzed. However, industrial experience has shown that such models, developed independently over the life cycle, result in multiple versions of the "truth" (i.e., they are not consistent with each other and the evolving architecture). The SAE Architecture Analysis and Design Language (AADL) standard addresses this issue of multiple truths due to inconsistency between analytical models by providing an architecture modeling notation with well-defined semantics that can accommodate multiple analysis dimensions through annotations and allow for auto-generation of these analytical models from a single source.The Carnegie Mellon Software Engineering Institute, L-3 Communications -EITS, and the Jet Propulsion Laboratory (JPL) have collaborated in a use of model-based engineering for the National Aeronautics and Space Administration (NASA) Software Assurance Research Program (SARP) project named "Model-Based Software Assurance with the SAE Architecture Analysis and Design Language (AADL)." The work involved applying the AADL to the Mission Data System (MDS) architecture. The SAE AADL industry standard for modeling and analysis of embedded software system architectures was chosen because of its ability to support analysis of nonfunctional properties, such as robustness, safety, performance, and security. The MDS was chosen because it takes an architecture-centric view by defining a multi-layered reference architecture for autonomous systems, whose dynamics are managed by feedback loops, and promotes state analysis through goal-oriented modeling to address uncertainty and faults. By combining the two technologies, we can take into account the impact of the embedded software's runtime architecture on these non-functional properties in the validation of systems.The result of that project sho...

No abstract

The ideas and findings in this report should not be construed as an official DoD position. It is published in the interest of scientific and technical information exchange.