Analysis of synchronized storage method for multimedia key areas based on machine learning

Abstract

Aiming at the problems of high energy consumption, small amount of stored data, large standard deviation of storage space, high failure rate of storage nodes and poor quality of data storage in existing multimedia key area synchronous storage methods, a machine learning based multimedia key area synchronous storage method is proposed. Using genetic algorithm to calculate the distance between data in multimedia critical area and cluster center, and redistribute cluster set, K-Mean is realized by M-R parallel computing model. Different amounts of data are allocated to the clustered sample data storage nodes to complete the synchronous storage of multimedia key area data. The experimental results show that compared with other methods, under the network regulation of 2800 m × 800 m, the storage energy consumption of the proposed method is in the range of 10 × 103NJ-1250 × 103NJ, and the storage energy consumption is low; The maximum number of data storage is 570, and the amount of stored data is large; the standard deviation of data storage space in the key area of multimedia changes within the range of 1.3–4, and the standard deviation of storage space is small. The proposed method lays a foundation for the further development of data storage technology.