A Survey of Service Identification Approaches - Classification Framework, State of the Art, and Comparison

Abstract

Due to their modular nature, the adoption of Service Oriented Architectures (SOA) for business applications promises many advantages. The successful introduction of SOA depends on an efficient methodical support of the underlying new development paradigm, though. As the amount of current literature illustrates, especially the development of systematic methods for the identification of suitable services, which can serve as building blocks of business applications, is presently a focal point of interest. The different approaches presented in literature, however, significantly diverge with regard to their concepts and procedures. In this paper, we therefore analyse the current state of the art in service identification and highlight differences between the presented approaches. To evaluate proposed service identification approaches, we introduce a classification scheme with distinguishing factors. We use this scheme to compare and analyse the various approaches. Based on this comparison, we elaborate on individual strengths and weaknesses of approaches from which implications for practice are deduced. Finally, we identify areas of future research that remain to be addressed in order to further advance the state of the art in service identification.