An architecture and coordination algorithms for improving the comprehensive performance of cloud storage

On the premise of in-depth study of cloud storage architecture and related factors affecting the overall performance of the system, a hierarchical control architecture of cloud storage system based on muti-master-slave structure is proposed by using the theory of Collaborative Artificial Intelligence. Secondly, taking two kinds of distributed file systems as the research object, the mathematical model and related formulas of the selection algorithm of distributed file system based on the architecture are given, and the selection results are analyzed and tested with a case. Finally, the performances of FastDFS, HDFS and the hierarchical architecture proposed in this paper are compared from various perspectives. The experimental results show that the overall performance of the architecture in cloud storage is better than that of a single system.