Power-Aware Virtual Machine Scheduling-policy for Virtualized Heterogeneous Multicore Systems

This paper presents a systematic approach to correctly provision server resources in data centers, resulting in minimum energy consumption and SLA violations. In particular, we describe a hybrid method for efficient server provisioning in virtualized heterogeneous multicore Cloud data centers. The objective is to place VMs on host while keeping total utilization of CPU below defined threshold and then optimizing the VM allocation by constrained consolidation of VMs leveraging live migration and switching off idle nodes to minimize power consumption. As excessive migration of VMs can lead to significant SLA violation which can hamper required Quality of Service, in proposed approach not only overall number of used servers are minimized, but also number of migrations are minimized, resulting in minimum energy consumption and least SLA violations. We present evaluation results showing that our approach can provide up to 13% of energy savings, reduced SLA violations and fewer VM migrations as compared to existing techniques, while preventing frequent power cycling of servers, thus justifying further development of stated policy. Keywords— Green IT; Cloud computing; Energy efficiency; Allocation of virtual machines; Resource management; Virtualization; Dynamic consolidation