UPB: User Preference based Brokering for Service Ranking and Selection in Federated Cloud

Cloud service selection in Federated Cloud Environment is a challenging task because of heterogeneity involved in service specifications, platforms, SLAs, etc. Users demand various QoS, which may be conflicting, for their applications. Cloud broker has to deal with the heterogeneity of federated cloud and users' demands. Cloud broker has to find a suitable service as per users' requirements from available services. Users also demand more than one services as per their requirements, so that they can choose anyone. Cloud broker has to find ranks of available services. The conflicting and non-commensurable demands from users make cloud service selection a multi-criteria decision-making problem. We have proposed User Preference based Brokering method to provide multi-criteria based service ranking and selection. Users request QoS parameters in some preference order as per their application requirement. The preference order is converted by assigning suitable weight to individual QoS parameters. The weighted QoS demands are evaluated with the available QoS values of every cloud services. The modified VIKOR method is applied in the next step to find the rank of services. The services get rank as per preference given to QoS parameters. The service which gets the highest rank is selected and allocated to user. The method is tested using CloudSim.