An Interface Complexity Measure for Component-based Software Systems

Controlling and minimizing software complexity is one of the most important objective of each software development paradigm because it affects all other software quality attributes like reusability, reliability, testability, maintainability etc. For this purpose, a number of software complexity measures have been reported to quantify different aspects of complexity. As the development of component-based software is rising, more and more complexity metrics are being developed for the same. In this paper, we have attempted to design an interface complexity metric for black-box components to quantify an important aspect of complexity of a component-based system. The proposed measure takes into account one major type of complexity of a component. It is due to its interactions (interfaces) with other components. Graph theoretic notions have been used to illustrate interaction among software components and to compute complexity. The proposed measure has been applied to five cases chosen for this study and yields quiet encouraging results which may further help in controlling the complexity of component-based systems so as to minimize both integration and maintenance efforts. As a thumb rule, we propose that average number of interactions (interfaces) per component in a component based system (CBS) should not be greater than five, otherwise that CBS would be highly complex and will be more prone to errors and hence unreliable. However, this rule requires further empirical support.