Service composition in dynamic environments: A systematic review and future directions

Abstract

Distributed computing paradigms such as cloud, mobile, Internet of Things, and Fog have enabled new modalities for building enterprise architectures through service composition. The fundamental premise is that the application can benefit from functionally equivalent services that can be traded in the cloud or services repositories. These services can vary in their Quality of Services (QoS) and cost provision. Accordingly, the problem of service composition is the process of choosing a configuration of candidate services from a pool of available ones, considering QoS attribute, cost, and users’ preference. Due to the inherent dynamism in service computing environments and communication networks, the advertised QoS values might fluctuate; therefore, service composition under uncertainty is inevitable and challenges satisfying Services Level Agreement (SLA). In this paper, we present a systematic literature review to investigate and classify the existing studies in service composition under uncertainty. We identified 100 relevant studies published between the year 2007 and 2020. To the best of our knowledge, this work is the first to explicate a focused systematic review, classification, taxonomy of approaches, and trends along with their assumptions and applications; and to discuss future research directions in service composition under uncertainty.