Behavioral Modeling of Software Intensive System Architectures

Architectural modeling and analysis are mechanisms that allow the capture of design decisions early in the process, so that they can be assessed and modified without incurring the costs of incorrect implementations. This paper addresses Monterey Phoenix (MP), a behavioral model for system and software architecture specification based on event traces, which supports architecture composition operations and views. MP captures behaviors and interactions between parts of the system and the environment with which it operates. As an executable architecture model leveraging “lightweight” formal methods and the small scope hypothesis, MP supports automatic generation of behavior examples (Use Cases) for early system architecture analysis, testing, verification, and validation. This paper also introduces a methodology utilizing MP that will inform quantifiable cost estimates (e.g. Function Point analysis) and ultimately project, program, and enterprise level resourcing decisions. Enhancing and extending DoDAF, UML, and SysML, MP is focused on behaviors, interactions, and automated tools for early verification.