Compatibility and performance analyses for component–based systems

As software engineering continues to adopt a component-based approach toward the construction of increasingly complex software architectures, the need to assess the compatibility and interoperability of the individual software components is becoming critical during the integration phase of the software production process [5], [9]. This assessment includes performance analysis of integrated systems and verification of temporal requirements which can be of primary importance for many real-time and embedded systems. While manual and ad hoc strategies toward component integration have met with some success in the past, such techniques do not lend themselves well to automation. A more formal approach toward the compatibility and interoperability assessment is needed. Such a formal approach would permit an assessment based on automated techniques and would also help promote the reuse of existing software components. Components represent high-level software models; they must be generic enough to work in a variety of contexts and in cooperation with other components, but they also must be specific enough to provide easy reuse. Although there are many (informal) component definitions [2], [7], very few attempts have been made to formalize these concepts. One aspect which many component definitions have in common is the notion of an interface that defines the component’s access points [8]. These access points allow other components to use the services provided by a component. Normally, a component can have multiple interfaces corresponding to its different access points. The behavior of a component, at its interface, can be represented by a (cyclic) labeled timed Petri net [4], [10]: