An improved genetic algorithm for the virtual machine placement problem

The dramatically increasing energy consumption of data centres is an important issue. Server consolidation is now regarded as one of the most effective ways to reduce the energy consumption in data centres. Server consolidation can be formulated as a Virtual Machine (VM) placement problem, and Genetic Algorithm (GAs) have been used to find an optimal solution to the VM placement problem. However, the slow computation time is a main concern of the GAs. This paper aims to improve the efficiency of the GAs. In this paper, we propose a new GA, namely Improved Genetic Algorithm (IGA), which is a GA enhanced by a new Initial Population Generation Strategy (IPGS), a new Knowledge-based Crossover Strategy (KCS), and a new Knowledge-based Mutation Strategy (KMS). Extensive experiments are conducted to illustrate the efficiency of the IGA.