Texture-Based Metrics Visualization on Software Architecture Diagrams

In this paper, we address the problem of visualizing several types of metrics computed on software architecture diagrams. Our specific aim is to present metrics computed on groups of diagram elements, such as classes or components in UML diagrams, together with metrics computed on diagram element members, such as class methods. For member metrics, we use an adapted version of the known table lens technique. For group metrics, we design a new solution that uses blended textures computed via spatial interpolation to show metric variations. Our method helps the task of visually correlating the distribution and outlier values of a multivariate metric dataset with the diagram structure. We present applications on architecture diagrams extracted from real-world software systems.