Redundant-free web services composition with user requirements consideration

• Main Author: Safi, Amirhossein
• Format: Thesis
• Language: English
• Published: 2013
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/33268/1/AmirhosseinSafiMFSKSM2013.pdf; http://eprints.utm.my/id/eprint/33268/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:72731?site_name=Restricted Repository

Service Oriented Computing (SOC) has become a leading study in recent years. It is based on the concept of composing distributed applications within heterogeneous environments by discovering and invoking web services to perform some complex tasks when there is no existing web service to fulfill the user request. Service Oriented Architecture (SOA) is known as a solution to accomplish composing autonomous, platform-free web services. However, the composition of web services without considering their underlying functional and non-functional attributes assures that it can adversely affect the composition process by increasing overall cost and leads to slow performance. This dissertation puts forward an idea of composing a redundant-free web service by considering user requirements. Non-functional attributes of services are prioritized to rank web services by using QoS aggregation function that leads to obtain more service requester satisfaction. After designing redundant-free web services composition, a method for optimal web services composition with several QoS attributes is proposed. At the end of this dissertation, the effectiveness and efficiency of proposed method based on execution time and optimal web service selection are evaluated. The proposed method is a considerable modification of the Global QoS Composition (GQC). This can be shown in the execution time improvement, which has been improved approximately seventeen percent as compared to the GQC method. In addition, the overall cost of web services composition is decreased due to deducting redundant Web services’ price from the sum of component services’ price in a composite service.