A Virtual Machine Placement Policy via Biogeography-based Optimization in the Cloud

Abstract

With the rapid development and popularization of cloud computing, cloud datacenters urgently need some highly efficient policies to improve their resource utilizations (e.g., power and physical machine resources). Although there are numerous virtual machine placement (VMP) policies which are designed to identify some most suitable physical machines for virtual machine requests, they are still not enough to improve their resource utilizations. Therefore, in this paper, we first exploit original biogeography-based optimization (BBO) algorithm to solve the VMP problem. In our BBO-based VMP policy, we first map the VMP problem to an ecosystem model, and then we redefine operators of the original BBO algorithm according to the ecosystem model. Finally, we perform a thorough performance evaluation of our BBO-based VMP policy in terms of the robustness, performance, and scalability. Experimental results show that our proposed policy outperforms other related policies.