Dynamic Resource Allocation Based on Energy Utility Maximization Using Virtual Machines in Cloud Environment.

Xiaohua Jia,Jinhai Wang,Chuanhe Huang,Qin Liu,Kai He,Jing Wang,Peng Li

2015-01-01

Abstract:With the popularity of cloud computing in recent years, rapid growth of the demand for computation power by scientific, business and web-applications has led to establishing plenty of large-scale data centers consuming enormous amount of electrical power. However, energy consumption has become an intractable problem. Recent advances in virtualization technologies have made it feasible to host multiple virtual machines (VMs) in the same physical host and even the same CPU core, with fair share of the physical resources among the VMs, and which makes laaS more scalable. In the context, We propose an energy efficient multi-dimension resource allocation algorithm for virtualized Cloud datacenters that reduces energy costs and provides required Quality of Service (QoS). Our VM deployment algorithm achieves a good balance between energy and performance by minimizing the amount of provisioning servers as well as maximizing time sharing of VMs hosted on the same server. Energy saving is achieved by VM deployment, continuous consolidation according to current utilization of resources, workload demand and load states of computing nodes. Our scheme achieves a good balance between energy consumption and performance. The dual-threshold strategy for VM migration efficiently reduces the times of VM migration and SLA violation. Meanwhile, we adopt dual-threshold DPS (dynamic powering on/off servers) techniques to power on/off servers and buffer the change of workload, and also adjust consolidation threshold dynamically. The results show that our proposed strategies bring sustainable energy saving while ensuring reliable QoS.