A novel virtual machine placement algorithm based on grey wolf optimization

Virtual machine placement is a typical NP-hard problem in the field of cloud computing. Unreasonable placement schemes can result in low resource utilization and high energy consumption. To address these issues, this paper proposes a heuristic algorithm named Adaptive Boundary Grey Wolf Optimization (ABGWO). The objective is to minimize the number of physical servers, and the evaluation is conducted using Microsoft’s Azure Trace dataset from 2020. Comparative experiments are performed with genetic algorithm, particle swarm optimization algorithm, artificial bee colony algorithm, moth search algorithm, and differential evolution algorithms. The experimental results demonstrate that the ABGWO algorithm outperforms other algorithms in terms of solution quality and stability.