FBM: A Fuzzy Brokering Model for Service Selection in Federated Cloud

Federated Cloud is made of multiple integrated cloud platforms governed by different cloud providers. The federated environment is created between various cloud providers either through agreement or standardization or third parties such as cloud brokers. The federation comprises services with varying access patterns, different performance levels, similar characteristics, and varying costs of use. These features create a heterogeneous environment that leads to a challenging service selection task. Cloud broker works as an intermediary between service providers and users to select the best service. A repository is created by a Cloud broker to store characteristics and users’ feedback on various parameters of services. A brokering model based on fuzzy logic has been proposed for ranking cloud services on both users’ feedback and characteristics. Based on the given criteria the best service is selected. The users provide feedback on various QoS parameters of a service that is used by them in linguistic terms. A new user requesting a service gives QoS requirements in linguistic form. It also provides a preference of demand for those QoS parameters. The linguistic terms can not be used in computation directly. The proposed model contains modules that convert these linguistic terms into numerical form. The model contains a method that considers both the preference of users for QoS parameters and feedback provided by other users in finding the rank of services. The proposed method is evaluated through simulation. Results are compared with existing literature. They show that the method successfully applies fuzzy logic in QoS modeling. The obtained results are compared on various performance metrics and they outperform than existing literature.