End-to-end QoS mapping and aggregation for selecting cloud services

In cloud computing environment, software service often collaborates with other supporting utility services in order to provide a complete solution to end users. The software service itself could provide the required functions to users that fulfill their needs. However, the non-functional requirements, which represent the main criteria in the service selection process, need the collaboration from all involved parties. Given the end-to-end non-functional requirements submitted from end users, the challenge is to select the best cloud solution that satisfies these requirements, and to let a software provider select the best utility services. In this paper, we propose a mapping mechanism that helps map users' QoS requirements to the required software service and utility services at different cloud levels. Ontology is used to semantically define diverging concepts of the non-functional requirements and guarantees and their relationships at different levels of cloud services. For implementation, OWL-S (Ontology Web Language for Services) and SWRL (Semantic Web Rule Language) are used. The experiment, in this paper, showed that the mapping process does not incur much calculation overhead, if integrated with any service selection system.