Automated Service Composition for On-the-Fly SOAs

In the service-oriented computing domain, the number of available software services steadily increased in recent years, favored by the rise of cloud computing with its attached delivery models like Software-as-a-Service (SaaS). To fully leverage the opportunities provided by these services for developing highly flexible and aligned SOA, integration of new services as well as the substitution of existing services must be simplified. As a consequence, approaches for automated and accurate service discovery and composition are needed. In this paper, we propose an automatic service composition approach as an extension to our earlier work on automatic service discovery. To ensure accurate results, it matches service requests and available offers based on their structural as well as behavioral aspects. Afterwards, possible service compositions are determined by composing service protocols through a composition strategy based on labeled transition systems.