Combining Local Optimization and Enumeration for QoS-aware Web Service Composition
Currently, many research efforts on Web services are made based on QoS attributes. However, user-side QoS data is not easily acquired since it relies heavily on user location and network status, and is dynamically changing from time to time. Aiming at measuring user-side QoS, the architecture of a QoS measure framework (QoS4WS) is introduced in this paper, which measures QoS for world-wide Web services with real invocations from different geographic regions periodically. When user-side QoS is to be acquired, a predication procedure which is based on a network coordination approach is executed to calculate real-time user-side QoS. With this framework, a more accurate user-side QoS can be measured automatically. Experiment results show that our framework can get a near-accurate user-side QoS.