Improving reliability in resource management through adaptive reinforcement learning for distributed systems

摘要

Demands on capacity of distributed systems (e.g., Grid and Cloud) play a crucial role in today's information era due to the growing scale of the systems. While the distributed systems provide a vast amount of computing power their reliability is often hard to be guaranteed. This paper presents effective resource management using adaptive reinforcement learning (RL) that focuses on improving successful execution with low computational complexity. The approach uses an emerging methodology of RL in conjunction with neural network to help a scheduler that effectively observes and adapts to dynamic changes in execution environments. The observation of environment at various learning stages that normalize by resource-aware availability and feedback-based scheduling significantly brings the environments closer to the optimal solutions. Our approach also solves a high computational complexity in RL system through on-demand information sharing. Results from our extensive simulations demonstrate the effectiveness of adaptive RL for improving system reliability.