Introducing test data-set for the QoS-aware web-services discovery and composition

Abstract

The QoS-aware web services discovery and composition are considered as two important, open and hot issues in Service-Oriented Architecture (SOA). By definition, web-service discovery is about how to select the best web-service for a role in a workflow among available web-services whereas web-services composition works on developing merely optimum coordination among a number of available web-services to provide a new composed web-service intended to satisfy some users requirements for which a single web service is not (good) enough. The criteria, upon which the web service selection, position, and composition processes are based, may or may not cover the Quality of services (QoS) parameters. The latter case would turn the name of the job into QoS-aware web services discovery and composition. In this article, the QoS-WSC test data-set is introduced for the QoS-aware web-services discovery and composition with regard to specific potentials and needs of the web-services world. In this respect, at first, an interface has been designed to define QoS for the roles attended in a service-oriented architecture. Then a solution, which allows comparison among web-services through the calculation of similarity of the request to amounts of the QoS parameters of available web services, is proposed. The similarity is obtained using the internal multiplication of two vectors of weighted numerical parameters representing request parameters and QoS parameters of available web services. The weights are technically significant coefficients, which can be obtained from the QoS-WSC data-set, which is assembled out of a rule-based integration of two well-known prior data sets in the field of web services.