A Dynamic Software Product Line Approach for Planning and Execution of Reconfigurations in Self-Adaptive Systems

Model-based autonomic computing systems facilitate the planning capabilities inside the adaptation logic. However, it is challenging to capture the complete reconfiguration behavior in a model. Context Feature Models used in Dynamic Software Product Lines help to specify the capabilities of a software as well as the monitored context values with the possibility to add constraints. Additionally, most adaptation logics are tailored to single use cases without the possibility for later reuse. This thesis presents an adaptation logic approach based on Dynamic Software Product Line variability models. The complete adaptation knowledge is encapsulated inside a knowledge component. This enables reuse of the complete adaptation logic. After the introduction of the approach, the adaptation logic is evaluated in a distributed computing use case