Abstract

Cloud computing has transformed modern computing infrastructure by providing scalable, on-demand access to computing resources. However, the rapid growth of cloud services has led to significant energy consumption, resulting in high operational costs and environmental impact. Efficient resource allocation in cloud environments is crucial to balance performance, energy efficiency, and service-level objectives. This paper investigates dynamic resource allocation algorithms designed for energy-efficient cloud infrastructure management. These algorithms adaptively allocate virtual machines (VMs), storage, and network resources based on real-time workload demands, system utilization, and energy consumption patterns. Techniques explored include heuristic-based scheduling, metaheuristic optimization, and machine learning-driven approaches, all aimed at minimizing power usage while maintaining quality of service (QoS). The study evaluates algorithm performance in terms of energy savings, task completion time, resource utilization, and SLA compliance across various cloud workloads. Experimental results demonstrate that dynamic allocation strategies significantly reduce energy consumption compared to static or naive allocation approaches while improving overall system efficiency. The paper highlights the importance of adaptive resource management policies for sustainable cloud computing, providing insights into the trade-offs between energy efficiency, computational performance, and service reliability in large-scale cloud infrastructures