基于密度演化的分布式大数据云存储方法仿真  

Simulation of Distributed Large Data Cloud Storage Method Based on Density Evolution

作　　者：徐丽[1] XU Li(School of Computer,Hubei University of Technology,Wuhan Hubei 430070,China)

机构地区：[1]湖北工业大学计算机学院,湖北武汉430070

出　　处：《计算机仿真》2021年第7期424-428,共5页Computer Simulation

基　　金：湖北省教育厅指导项目(B2015051)。

摘　　要：针对传统分布式大数据存储方法效率慢、能耗高、容易出现冗余信息的问题,提出一种基于密度演化的分布式大数据云存储方法。采用随机系统分析密度演化基础理论,确定相关参数值和关联规则,据此建立分布式云存储基础模型,保证大数据传输存储过程中每条数据能够独立存在,并增强冗余数据的分配准确率和分配效率,通过计算获得云存储中冗余数度分类,构建最佳冗余数据分配策略,最大程度将低存储时间和空间。通过计算数据融合概率实现存储压缩,得出大数据的变化特征识别数学模型,由此实现密度演化的分布式大数据云存储。仿真结果表明,所提方法能够有效减少冗余信息对云存储的影响,且整个过程耗时短、耗能低、效率高,可广泛使用在现实环境中。Aiming at the problems of low efficiency, high energy consumption and easy to appear redundant information in traditional distributed big data storage methods, a distributed big data cloud storage method based on density evolution is proposed. The stochastic system was used to analyze the basic theory of density evolution and determine the relevant parameter values and association rules. On this basis, a basic model of distributed cloud storage was built to ensure that each data could exist independently in during the big data transmission and big data storage, so that the allocation accuracy and allocation efficiency of redundant data was enhanced. Through the calculation, the classification of redundant data in cloud storage was obtained and the optimal redundant data allocation strategy was constructed, so that the storage time and space were reduced to the most extent. The storage compression was realized by calculating the data fusion probability, and the mathematical model which recognizes the change characteristic of big data was built. Finally, the distributed big data cloud storage based on density evolution was completed. Simulation results show that the proposed method can effectively reduce the influence of redundant information on cloud storage. Meanwhile, the whole process requires less time and lower energy consumption, so this method can be widely applied in real-world environments.

关 键 词：密度演化 云存储 冗余数据分配 分布式大数据 数据存储方法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]