Regression Tree based Ranking Model in Federated Cloud

Background/Objectives: Federated cloud architecture is heterogeneous and distributed model that provides infrastructures related to cloud by aggregating different IaaS providers. In this position, it is an exciting task to pick the optimal cloud provider for the services and deploy it in affordable cost. In this paper, an effective new methodology for establishing trust and the different cloud providers based on the trust value using SMI attributes based on regression tree model is to enrich security and privacy in the federated cloud. Methods/Statistical Analysis: The scoring is computed based on the depth of the traversals in regression tree. Policy based trust, SLA verification trust; reputation trust and Evidence based trust are proposed in the architecture to build secured federated cloud. Findings/Conclusion: Broker trust level can also be identified and classified into four different categories such as completely trusted, partially trusted, minimally trusted and not trusted. Simulation results show that the performance of the ranked based mechanism depends on its implementation and better than without rank based federated cloud. Applications/Improvements: Will focus on mathematically formal frameworks for reasoning about trust, including modeling, languages and algorithms for computing trust.