FULL PDF
Abstract
Modern data centersare undergoing a transformative shift as they adapt to dynamic workloads spanning heterogeneous platforms such as edge, fog, and cloud systems. Artificial Intelligence (AI)-driven workload optimizers are emerging as critical tools for dynamic resource allocation, enabling energy efficiency, cost reduction, and improved quality of service (QoS). This paper introduces a multilayered AI workload optimization framework tailored for cross-platform data centers. The framework incorporates real-time decision-making, predictive modeling, and reinforcement learning mechanisms to allocate resources adaptively across distributed infrastructures. By drawing upon advancements, we assess the challenges and performance implications of dynamic workload optimization, culminating in a hybrid AI system that balances scalability and responsiveness.
View more >>