A Field-Based Model for Representing Dynamic and Evolving Features of Cloud Services

In cloud computing context services present the dynamic and evolving features, which greatly affects the accuracy and efficiency of service discovery. It is the prerequisite for supporting service discovery to capture and represent these features during the process of services organization and management. This paper proposes a field-based model to describe the dynamic and evolving features of cloud services in both qualitative and quantitative way, which is inspired by the Bohr atom model. The concept of energy level in Bohr model is used to represent the services status and demarcate cloud services, electrons jumping mechanism in Bohr model is used to depict services' dynamic and evolving features and analyze how services status are changed from one energy level to another. The field model of services provides abstractions to classify a set of services according to their status and mechanism to explain their changes and demarcation according to their potential energy variation. The concept of user acceptable services region is proposed to represent the search scope and method of user service discovery request in cloud services field model. The algorithms to generate field model of services and form user acceptable services region are designed, with which field-based service discovery algorithm is proposed. Based on QWS dataset, we conduct a series of experiments to evaluate and validate the effectiveness of field-based service model for organizing and discovering cloud services.