云平台多密级数据分区存储节点选择方法  

The Method of Selecting Storage Nodes in Multi Secret Data Partition of Cloud Platform

作　　者：黄正鹏 周元哲 娄必伟 贺道德 HUANG Zheng-peng;ZHOU Yuan-zhe;LOU Bi-wei;HE Dao-de(College of Information Engineering,Guizhou University of Engineering Science,Bijie Guizhou 551700,China;School of Computer Science,Xi`an University of Posts&Telecommunications,Xian Shanxi 710121,China)

机构地区：[1]贵州工程应用技术学院信息工程学院,贵州毕节551700 [2]西安邮电大学计算机学院,陕西西安710121

出　　处：《计算机仿真》2020年第11期375-379,共5页Computer Simulation

基　　金：贵州省教育厅青年科技人才成长项目(黔教合KY字[2016]289);贵州省教育厅创新群体重大研究项目(黔教合KY字[2016]057)。

摘　　要：数据储存节点选择方法易忽略数据的重复性和权重指标,由此产生较多冗余数据,增加节点能量消耗。基于此提出云平台多密级数据分区存储节点选择方法。结合云储存的多副本理念对网络区域编码,并剔除冗余编码;以云平台分区储存系统原理与数据包生成过程为储存节点选择的依据,确定节点选择的四项权重因子;结合权重指标将节点选择问题转化为统筹领域中的多属性决策问题,计算归一化处理的原始数据矩阵中决策问题的正理想解与负理想解之间的差距,以此为评价基础,选择分区储存的最佳节点位置。仿真结果证明,节点能量消耗较小,提升了云平台储存容量,能够实现节点的负载动态调节。Traditionally,the method is easy to ignore the repeatability and weight index of data,resulting in the increase of redundant data and energy consumption of nodes.On this basis,this paper proposed a method to select partition storage nodes in multi secret data of cloud platform.Combined with the concept of multiple copies,the network region was coded and then the redundant codes were eliminated.The principle of partition storage system and the process of data packet generation were taken as the basis for selection of storage nodes.After that,four weight factors of node selection were determined.Combined with the weight index,the problem of node selection was transformed into the problem of multi-attribute decision in the field of overall planning.Finally,the difference between the positive ideal solution and negative ideal solution of the decision-making problem in the original data matrix after normalization was calculated.Thus,the location of best node in partition storage was selected.Simulation results prove that the energy consumption of node is small,and the storage capacity of cloud platform is improved,so this method can adjust the load of nodes can be adjusted dynamically.

关 键 词：云平台 多密级数据 分区储存 节点选择 权重指标 

分 类 号：TP390[自动化与计算机技术—计算机应用技术]