Remote Sensing Data Processing Process Scheduling Based on Reinforcement Learning in Cloud Environment

Task scheduling plays a crucial role in cloud computing and is key factor determining performance. To solve the task problem for remote sensing data processing computing, this paper proposes workflow algorithm---Workflow Scheduling Algorithm based on Deep Reinforcement Learning (WDRL). The process modeling transformed into directed acyclic graph problem. Then, algorithm designed by establishing Markov decision model adopting fitness calculation method. Finally, combine advantages of reinforcement learning deep neural networks to minimize make-time processes from experience. experiment development CloudSim Python compares change completion time data. results show that compared with several traditional meta-heuristic algorithms, WDRL can effectively achieve goal optimizing efficiency.