Using Dependence Matrix to Support Change Impact Analysis for CBS

Abstract

Component-based software development technique and its extensive application have led to the wide research in various aspects of component-based software (CBS). The rapid evolution of CBS brings great challenges to its maintenance in the later phase, so it is quite necessary to measure the change impact on the whole system. By using component dependence matrix to represent component-based software system, the paper mainly discusses the case of component modification, including the single component change and the changes of multi-components, and proposes the corresponding algorithms (i.e. Appxm_RM, Generate_SP, and Calculate_CR) for analyzing the change impacts in CBS. In addition, the calculation process of change impact analysis and its effectiveness are also validated by a simple CBS.

Dependency and interaction oriented complexity metrics of component-based systems

“…Thus, coupling is considered as a measure of intercomponent dependency. It is widely accepted that excessive coupling is not a good design practice and it usually results in complex system that are difficult to maintain and update [9]- [15]. This metric computes the ratio of total number of direct interactions between the components to total number of components.…”

A practice-driven systematic review of dependency analysis solutions

“…Dependency analysis support change impact analysis in many different ways (see Table 3). First, some solutions are presented to support change impact analysis on specific type of systems, e.g., aspect-oriented (Zhao 2002), component-based (Mao et al 2007), and object oriented (Huang and Song 2007;Xing and Stroulia 2006). Second, many solutions are for supporting activities that facilitate change impact analysis, e.g., change prediction (Hassan and Holt 2004;Kagdi and Maletic 2007;Law and Rothermel 2003b), identification of dependence clusters and dependence pollution (Binkley and Harman 2005), dynamic impact analysis in object-oriented programs (Huang and Song 2007), identification of class change profiles (Xing and Stroulia 2006), impact of database schema change (Maule et al 2008), and efficient source code navigation (Robillard 2008).…”

“…Dependency analysis solution work with various types of diagrammatic representations such as top-down descriptions (McComb et al 2002), component-based models (Vieira and Richardson 2002), matrix models (Khan et al 2008;Mao et al 2007;Xing and Stroulia 2006), chart diagrams (Garousi et al 2006;Moraes et al 2005;Ryser and Glinz 2000), and business process models (Ivkovic and Kontogiannis 2006;Vasilache and Tanaka 2005). Top-down descriptions are used to construct dependency models for application level analysis (McComb et al 2002).…”

“…interfaces and ports but using semi-formal interface specifications (Vieira et al 2001). Matrix models are the basis for the construction of dependency matrices to analyze changes in UML models and detect design-level structural modification (Khan et al 2008;Mao et al 2007;Xing and Stroulia 2006).…”