Currently a lot of Web-services are used in variety of business processes. There are many Web-services that have the similar functionalities but differ by Quality of Service (QoS). As such, the problem becomes how to select appropriate web services that best suited to customer requirements, like minimum cost, lowest response time, maximum throughput, etc. In this paper we analyze the publications which related with this problem, and surveys on classification approaches of its decision, as well as the choice of categories to classify these approaches.
The set of the QoS requirements used in approaches, methods and algorithms for discovery, selection and negotiation, varies considerably. OASIS consortium proposed to divide the quality factors of Web-services into two groups: the business quality group and the system quality group. System quality group is comprised of the variant quality part and the invariant quality part. Variant quality part takes into account the consumer's characteristics of Web-service, while the invariant part includes technical characteristics.
The methods for discovery and matching of Web-services are grouped according to various criteria: static and dynamic, the using of software agents, the including of third parties for process harmonization consumer's requirements and provider's proposals, and so on.
Analysis of approaches for discovery and matching Web-services has shown that they can be divided into three main areas. First, the approaches based on syntactic description of Web-service in formal languages like WSDL, WS-BPEL, API UDDI and other. This approaches limited for correct interpretation of QoS-properties. Second, the approaches based on semantics, common and domain ontologies for identical understanding of QoS concepts by consumer and provider of Web-service. Third, the context-aware approaches that take account of conditions and states of consumer in runtime of searching and request Web-service.
These results of the analysis can be used to identify promising areas of research and development QoS-aware approaches, models and methods for discovery and matching of Web-services.