Energy-Efficient Cloud Systems with VM Consolidation Optimization

This technical report presents a study focused on improving resource utilization and energy efficiency in cloud computing through virtual machine (VM) consolidation. The authors introduce a mixed integer linear programming (MILP) model that incorporates G-robustness theory to address uncertainties in VM usage, aiming to optimize both performance and energy consumption. A heuristic algorithm is developed to facilitate large-scale VM allocation. The research includes experiments conducted using data from Huawei Cloud, which demonstrate significant enhancements in resource utilization and energy efficiency. The report outlines the challenges of VM placement in cloud environments, emphasizing the need for effective resource management strategies to minimize operational costs and energy consumption. It also discusses the complexities of dynamic consolidation, including the balance between maximizing resource utilization and maintaining quality of service. The findings indicate that the proposed model and algorithm can effectively consolidate VMs, showcasing practical applicability in real-world cloud settings.