Trust Based VM Consolidation in Cloud Data Centers

Abstract

Virtualization in Cloud data center, handle workloads and maintain SLA providing a better QoS to the Cloud consumer will lead to the harnessing of the present Cloud Computing infrastructure. Our model is on a statistical property and based on reliability and reputation combined for a "trust" based that we design our algorithms to handle QoS and these algorithms prove better than the existing model. However, the growing demand of the resources (physical) in a data center has drastically increased the energy consumption of computations (cyber) being processed in data centers, which has become a decisive issue. To address the trade-off between performance and power consumption we propose a near-optimal scheduling policy based on the CQR (Composite Quantile Regression) and the Minimum energy heuristics (MPP) to find a trust based Cloud character probability modeling that exploits heterogeneity across multiple data centers for a Cloud provider. © Springer-Verlag Berlin Heidelberg 2014.