An Integrated Model-Driven Development Environment for Composing and Validating Distributed Real-Time and Embedded Systems

Summary

Model-driven development (MDD) tools and processes are increasingly used to develop component middleware and applications for distributed real-time and embedded (DRE) systems, which have stringent requirements for timeliness, correctness, scalability, and maintainability. MDD techniques help developers of DRE systems express application functionality and quality of service (QoS) requirements at a higher level of abstraction than is possible using third-generation programming languages, such as Visual Basic, Java, C++, or C#. The state-of-the-art in MDD for large-scale DRE systems is still maturing, however, and no single MDD environment provides the capabilities needed for effective development of large-scale DRE systems.

This chapter presents three contributions to the study of integrated MDD development and model checking for large-scale DRE systems. First, we describe how our CoSMIC and Cadena MDD toolsuites have been combined to provide an integrated environment that enhances the development and validation of DRE systems. Second, we discuss how we addressed key research issues associated with implementing MDD algorithms for maintaining semantics-preserving transfer of model data between the CoSMIC and Cadena MDD tools. Third, we discuss how we overcame technical difficulties encountered when applying the integrated COSMIC and Cadena for a representative DRE system. Our results show that interoperation between different MDD tools is achievable with the proper choice of communication format, semantics, and the development of a reliable graph diff-merge algorithm. This interoperation helps identify the workflow and capabilities needed for next-generation DRE development environments.

This work was sponsored in part by NSF ITR #CCR-0325274, NSF ITR #CCR-0312859, DARPA/AFRL #F33615-03-C-4112, and Lockheed Martin.