Effects of Correlation-Based VM Allocation Criteria to Cloud Data Centers
Jing V. Wang, Chi‐Tsun Cheng, Chi Kong Tse · 2016
Virtualization technology has been widely adoptedin Cloud data centers for adaptive resource provisioning. Withvirtualization, multiple virtual machines (VMs) can be colocatedon a single physical host to yield maximum efficiency. However, VMs which show high CPU utilization correlationsto other co-located peers are more likely to trigger overloadingincidents. This work provides an analysis on effects ofcorrelation-based VM allocation criteria to Cloud data centers. The correlations among VMs' CPU utilizations are consideredas parameters for decision making in VM allocation processes. Three different expressions of correlation-based criteria areintroduced and evaluated in this work. According to oursimulation results obtained from CloudSim with real-worldworkload traces, Cloud data centers with correlation-based allocationcriteria can perform better in terms of reducing energyconsumption and avoid committing Service Level Agreementsviolations than those with power-based criteria.