Green Cloud Computing: Energy-Efficient Strategies for Sustainable Data Centres

Abstract

The rapid expansion of cloud computing, artificial intelligence, Internet of Things applications, streaming platforms and digital services has significantly increased the demand for data-centre infrastructure. Although cloud computing provides scalable and flexible computational resources, the associated growth in servers, storage systems, cooling infrastructure and networking equipment has created substantial energy and environmental challenges. Green Cloud Computing has emerged as an important research direction within computer science that seeks to reduce the energy consumption and environmental footprint of cloud infrastructures while maintaining performance, reliability and service quality. This paper examines the conceptual foundations of Green Cloud Computing and evaluates major energy-efficient strategies for developing sustainable data centres. Particular attention is given to server virtualization, workload consolidation, dynamic resource allocation, energy-aware scheduling, efficient cooling, renewable-energy integration, workload migration, energy-efficient networking and carbon-aware cloud computing. The paper also discusses the role of artificial intelligence and machine learning in predicting workloads, optimizing resource utilization and managing data-centre energy demand. The relationship between energy efficiency, performance, service-level agreements and carbon emissions is critically examined. Furthermore, the paper identifies challenges related to heterogeneous hardware, workload variability, renewable-energy intermittency, cooling requirements, embodied carbon and the growing computational demands of artificial intelligence. The analysis suggests that sustainable cloud computing requires a holistic approach rather than isolated energy-saving techniques. Future green data centres are likely to combine intelligent resource management, renewable energy, advanced cooling technologies, efficient hardware and carbon-aware workload orchestration.