Tree-based Search Algorithm for Web Service Composition in SaaS

With the growing demand of cloud computing most companies are moving software to the cloud in form of web services. Web services are popular in terms of distributed technology that can successfully solve integration problems between heterogeneous systems. Web service composition is NP-hard problem and one of the most challenging problems in web services. This paper aims to optimize QoS attributes in web service composition using a tree-based heuristic search algorithm. The proposed algorithm consists of three main steps: tree creation, tree filtering and tree searching. First the algorithm creates a tree that represents all possible composition solutions. Second the filtering algorithm removes illegal composition paths as well as composition paths that less likely to be in the optimal set. Finally a best-first search algorithm is used to find, evaluate and rank the optimal composition solutions. To demonstrate the efficiency and effectiveness of the proposed algorithm, we provided experimental results for supply chain management system case study with multi QoS attributes.