Models@ Run.time to Support Dynamic Adaptation

Today´s society increasingly depends on software systems deployed in large companies, banks, airports, and so on. These systems must be available 24/7 and continuously adapt to varying environmental conditions and requirements. Such dynamically adaptive systems exhibit degrees of variability that depend on user needs and runtime fluctuations in their contexts. The paper presents an approach for specifying and executing dynamically adaptive software systems that combines model-driven and aspect-oriented techniques to help engineers tame the complexity of such systems while offering a high degree of automation and validation.