Dynamic Multi Granularity Service Composition

The trend is for enterprises to outsource parts of their services, in order to concentrate on their own core `businesses. Meanwhile, users usually need to compose multiple different services to create a sophisticated application. Through the service-oriented architecture paradigm, users can compose elementary services to form new value added services through the process of service composition. In template-based service composition, an abstract composite service, consisting of a collection of abstract services orchestrated by workflow patterns, is first defined and then instantiated and executed at run time by binding abstract services to concrete ones. This dynamic binding ensures a loose -coupling of services and all so-called QoS-aware service composition problem. In existing work, to expand the selection scope using the concept of generalized component services, a backtracking-based algorithm and an extended genetic algorithm(GA) has been applied for finding an optimized solution and near-optimal solution respectively in composition service The proposed work, will adopt the multi-granularity service composition automatically at run time. This will be useful to study how to extend other Meta-heuristic algorithms along with Tabu-search algorithm used for efficient optimization service selection.