Variability middleware for multi-tenant SaaS applications: a research roadmap for service lines

Abstract

Software product line engineering (SPLE) and variability enforcement techniques have been applied to run-time adaptive systems for quite some years, also in the context of multi-tenant Software-as-a-Service (SaaS) applications. The focus has been mainly on (1) the pre-deployment phases of the development life cycle and (2) fine-grained (tenant-level), run-time activation of specific variants. However, with upcoming trends such as DevOps and continuous delivery and deployment, operational aspects become increasingly important. In this paper, we present our integrated vision on the positive interplay between SPLE and adaptive middleware for multi-tenant SaaS applications, focusing on the operational aspects of running and maintaining a successful SaaS offering. This vision, called Service Lines, is based on and motivated by our experience and frequent interactions with a number of Belgian SaaS providers. We concretely highlight and motivate a number of operational use cases that require advanced variability support in middleware and have promising added value for the economic feasibility of SaaS offerings. In addition, we provide a gap analysis of what is currently lacking from the perspectives of variability modeling and management techniques and middleware support, and as such sketch a concrete roadmap for continued research in this area.