Claims and Evidence for Architecture-based Self-adaptation: A Systematic Literature Review

Abstract. Engineering the upcoming generation of software systems and guar-anteeing the required qualities is complex due to the inherent uncertainties at design time, such as new user needs and changing availability of resources. Architecture-based self-adaptation is a promising approach to tackle these chal-lenges. In this approach, a system maintains a model of itself and adapts itself to realize particular quality objectives using a feedback loop. Despite a vast body of work, no systematic study has been performed on the claims associated with architecture-based self-adaptation and the evidence that exists for these claims. As such insight is important for researchers and engineers, we performed a sys-tematic literature review covering 20 leading software engineering conferences and journals in the field, resulting in 121 studies used for data collection. The review shows that self-adaptation is primarily used to improve performance, reli-ability, and flexibility. The tradeoffs implied by self-adaptation have not received much attention, and evidence is mainly obtained from simple examples. From the study, we derive a number of recommendations for future research in architecture-based self-adaptive systems.