Enhancing Kubernetes automated scheduling with deep learning and reinforcement techniques for large-scale cloud computing optimization

Abstract

These artificial intelligence technologies have become key tools for solving automated task scheduling as cloud computing applications continue to expand. The paper proposes an automatic task scheduling scheme for large-scale cloud computing systems based on deep learning and reinforcement learning. The scheme addresses the complexity and real-time requirements of task scheduling in such systems. The cloud computing system's parameters are monitored and predicted in real-time using deep learning technology to obtain system status information. The task scheduling strategy is then dynamically adjusted based on the real-time system state and task characteristics, using a reinforcement learning algorithm to achieve optimal utilization of system resources and maximum task execution efficiency. This paper examines the effectiveness and performance advantages of the proposed scheme through experiments, demonstrating the potential and application prospects of deep learning and reinforcement learning in automatic task scheduling for large-scale cloud computing systems.