No abstract

No abstract

To improve resource utilization and energy efficiency, cloud datacenters use VM consolidation to consolidate VMs to less number of physical machines through VM migration. However, improper VM placement may cause frequent VM migrations and constant on-off switch on physical machines (PMs), which results in decreasing service quality and increasing energy consumption. To address this problem, in this paper, we propose an effective and efficient VM consolidation approach called EQ-VMC with the goal to optimize energy efficiency and service quality. In our approach, a discrete differential evolution algorithm is developed to search the global optimum solution of VM placement. By integrating it with a set of algorithms we propose for effective host overloading detection, VM selection and under-loaded host detection, EQ-VMC effectively reduces energy consumption and improves quality of services (QoS). Extensive simulation demonstrates its effectiveness and shows its advantage compared with previous VM consolidation methods.

In this paper, we develop a model for the competitive relationship between cloud service providers and cloud brokers during the virtual machine (VM) placement. The model adopts the idea of a strategy game, in which the cloud service provider and all VMs are participants. First, we define the strategy for the participants in the strategy game model according to the difference between the number of physical resources provided by the provider, and that requested for the VMs by the brokers. Next, payment values between the providers and brokers are obtained from each strategy in the game matrix, which corresponds to the strategy set of VM placement. In addition, the Nash equilibrium solution of the current strategy game demonstrates the trade-off between reducing energy consumption and guaranteeing QoS. On the basis of this, the optimal VM placement scheme is obtained, and the ongoing VM consolidation is examined. Finally, a strategy-game-based VM consolidation (SG-VMC) method is proposed. The extensive experimental results show that the proposed SG-VMC method can significantly reduce energy consumption while guaranteeing QoS.INDEX TERMS Game theory, strategy game model, VM placement, VM consolidation.

No abstract