ITR: Principles of Distributed Component-Based Software

Ohio State University Research Foundation -Do Not Use, Columbus OH

Investigators

Abstract

Reasoning about the behavior of large component-based software systems demands a "modular" or "compositional" reasoning system, in which summary properties of a system's pieces are composable to deduce properties of the entire system without delving into the internal details of those pieces. This research focuses on contributing principles for how to design component-based software that supports modular reasoning, and to help bring this new knowledge into practical application with commercial distributed component technologies. Specifically, the project investigates: (1) developing and describing detailed principles for designing the interfaces of software components so that they support both modular reasoning about system behavior and effective and efficient distribution and execution; (2) showing how to write human-understandable behavioral specifications for the interfaces of components designed using the above principles; (3) demonstrating additional practical benefits from having formal specifications available to software engineering tools. The generality and efficacy of the results will be evaluated through construction of prototype tools that support distributed component-based software design and development in a programming-language-neutral environment, and by observing the effects of using such tools in the classroom to see how much students benefit from the tools' new specification-enabled capabilities as they design and develop distributed component-based software systems.