Research Article
Research on Normalized Network Information Storage Method Based on Deep Reinforcement Learning
@INPROCEEDINGS{10.1007/978-3-030-82562-1_56, author={Qiang Wang and Lai-feng Tang}, title={Research on Normalized Network Information Storage Method Based on Deep Reinforcement Learning}, proceedings={Multimedia Technology and Enhanced Learning. Third EAI International Conference, ICMTEL 2021, Virtual Event, April 8--9, 2021, Proceedings, Part I}, proceedings_a={ICMTEL}, year={2021}, month={7}, keywords={Deep reinforcement learning Normalized network Information storage Hash function Archived data}, doi={10.1007/978-3-030-82562-1_56} }- Qiang Wang
Lai-feng Tang
Year: 2021
Research on Normalized Network Information Storage Method Based on Deep Reinforcement Learning
ICMTEL
Springer
DOI: 10.1007/978-3-030-82562-1_56
Abstract
This paper makes a deep research on the standardized network information storage method, applies deep reinforcement learning to the standardized network information storage, and proposes a standardized network information storage method based on deep reinforcement learning. Each file to be archived is firstly divided into non-overlapping semantic fragment data blocks according to its information content. Each data block will encrypt its content through hash function, and obtain a signature as its identifier for data archiving. The archived data is divided into task data, resource data and document data. The rack-aware data placement strategy was developed based on the deep reinforcement learning algorithm to improve data reliability, availability and network bandwidth utilization, and the cloud storage model was designed to normalize the network information storage of classified archived data. Experiments show that this method has a high average disk read and write speed and can meet the requirements.
