|Authors||R. Behjati, T. Yue and L. C. Briand|
|Title||A Model-Based Approach to the Automated Reuse of Configuration Data Based on Internal Similarities|
|Publication Type||Technical reports|
|Year of Publication||2012|
|Publisher||Simula Research Laboratory|
Product configuration in families of embedded software systems, such as Integrated Control Systems (ICSs), involves resolving thousands of configurable parameters and is, therefore, time-consuming and error-prone. Typically, these systems consist of highly similar components that need to be configured similarly. For large-scale systems, a considerable portion of the configuration data can be reused, based on such similarities, during the configuration of each individual product. In this paper, we propose a model-based approach to automate the reuse of configuration data based on the similarities within an ICS product. Our approach provides configuration engineers with appropriate means for manipulating the reuse of configuration data, and provides the required formalism for ensuring the consistency of the reused data. Our investigation of a number of product configurations with an industry partner shows that more than 60% of configuration data can be automatically reused using our similarity-based approach, thereby reducing configuration effort.