Joint Load-Balancing and Energy-Aware Virtual Machine Placement for Network-on-Chip Systems

Abstract

Virtualization is one of the key enabler technologies of cloud computing in providing on-demand sharing of computing resources. Virtualization requires mechanisms and algorithms for virtual resource allocation, virtual machine deployment, migration, and servers consolidation. Most of the existing studies have only focused on how to solve the problem of virtual resource allocation among servers. However, as cloud servers with multi-core architectures become popular, the virtual machine resource allocation in a single server becomes a critical challenge. In this paper, we propose a multi-objective virtual machine placement algorithm by jointly considering energy efficiency and load balancing criteria in a multi-core server with the Network-on-Chip architecture. Our proposed algorithm is based on Markov approximation optimization theory. We perform extensive experiments to evaluate our proposed algorithm. The results show that our proposed algorithm achieves higher energy efficiency, load balancing, and calculation speed compared with the state-of-the-art algorithms.