AI-Powered Smart Energy Management for Optimizing Energy Efficiency in High-Performance Computing Systems
M. Jalasri, Soumyashree M. Panchal, Karpagam Mahalingam, Rama Venkatasubramanian, R. J. Hemalatha, Sampath Boopathi · Advances in computational intelligence and robotics book series · 2024
The chapter discusses the need for efficient energy consumption in high-performance computing systems and proposes the integration of artificial intelligence and machine learning techniques to optimize energy efficiency. It explores AI-driven techniques like reinforcement learning, neural networks, and predictive analytics for energy-aware scheduling, workload allocation, and adaptive power management. The chapter discusses the effectiveness of AI-driven energy optimization strategies in real-world HPC infrastructures, highlighting potential energy savings while maintaining computational performance. It also discusses future directions and challenges in AI-enabled smart energy management, including algorithm refinement, integration with emerging technologies, and scalability considerations. The holistic approach highlights the transformative impact of AI and ML in creating sustainable, energy-efficient paradigms within high-performance computing ecosystems.