Modelling and Analysis of Feature-Oriented Software

点击翻译按钮获取中文摘要

Software systems are growing in size and complexity. The "scale" of large-scale systems no longer refers simply to large codebases but also to extreme variability, as characterized by features. Features are particularly relevant in software product lines, in which a family of related software products (e.g., smart phones, automotive software) shares a common set of mandatory features and products are differentiated by their variable (optional or alternative). When deriving a product from a selection of features, software developers must consider how the features interact. Some interactions are planned, some are innocuous, and others are harmful. For example, the software controllers for the braking features on the 2010 Toyota Prius interacted badly, reducing drivers' ability to brake and leading to 62 reported crashes and 12 injuries. To be safe, software developers analyze the consequences of all possible feature interactions, in order to find and fix the bad interactions. The Feature Interaction Problem is that the number of interactions to consider is exponential in the number of features. As a result, a software team finds that its work in developing new features is dominated by the tasks to detect, analyze, and verify interactions. A second problem is that software developers do not realize the full potential of feature-oriented development because verification is typically performed on derived products and it is too costly to verify all of the possible products. Instead, a small fraction of products are derived, verified and offered to customers.