In a multi-tenant service network, services relate to each other and collaborate to support the functional and performance requirements of multiple tenants. Such a service network evolves over time as its services and tenants change. Consequentially, the composite application that enacts the service network also needs to evolve at runtime, which is problematic. For example, different types of changes to the application, and their consequential impacts need to be realized and managed at runtime. In this paper, we present an approach to evolving multi-tenant service networks. We identify the types of runtime changes to a service network composite application and their impacts, and present a middleware support for realizing and managing the identified changes and impacts. A software engineer can specify the desired changes to the running application, and enact the change specification to modify it. We show the feasibility of our approach with a detailed case study.
CITATION STYLE
Kumara, I., Han, J., Colman, A., Van den Heuvel, W. J., & Tamburri, D. A. (2018). Runtime evolution of multi-tenant service networks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11116 LNCS, pp. 33–48). Springer Verlag. https://doi.org/10.1007/978-3-319-99819-0_3
Mendeley helps you to discover research relevant for your work.