AI-Powered Resource Allocation for Dynamic Cloud Workloads

To tackle the complexities of dynamic and fluctuating workloads in cloud computing, this paper introduces a novel resource allocation approach leveraging Transformer-based models and reinforcement learning techniques like Proximal Policy Optimization (PPO) and Deep Q-Networks (DQN). The proposed framework dynamically coordinates resource distribution among virtual machines, adapting to workload variations and task durations. This strategy significantly improves resource reuse, maximizes data center utilization, and reduces operational costs. The results underscore the effectiveness of these AI-driven techniques in achieving robust and adaptive cloud resource management.