摘要
对分布式数据存储过程的研究,能够有效提高云计算环境下数据处理效率。对云计算中分布式数据的存储,需要设计了存储数据包分类模型,精确划分存储数据块各个区域,完成分布式数据的分区存储。传统方法结合蛇形时隙数据存储方法,确定数据存储位置至查询节点网格的距离,但忽略了对待存储数据进行高效分区,导致存储效果不理想。提出云计算中分布式数据存储方法。将云计算中分布式海量源数据包依据一定接收概率,分散至分布式系统中的全部节点,在各个节点构成一个存储数据包。设计存储数据包分类模型,将存储区域划分为热、冷数据块区以及重复存储数据包区,依据活动因子特点进行分区存储,完成分布式数据的高效存储。实验结果表明,上述方法能够有效提升系统磁盘存储容量占用率,实现了数据节点负载动态调节。
This article proposes a method for distributed data storage in cloud computing. Our research dispersed distributed mass source data packet in cloud computing to all nodes in distributed system according to certain proba- bility of acceptance, then formed a storage data packet in each node. The research designed classification model of the storage data packet and divided storage area into cold data block area, hot data block area and data packet area of repetitive storage. According to feature of active factor, the research carried out partition storage. Thus, we comple- ted the distributed data storage efficiently. Simulation results show that the method can improve occupancy rate of storage capacity of system disk and achieves dynamic regulation of node load of data.
作者
侯桂云
陈桂英
卢志强
HOU Gui-yun;CHEN Gui-ying;LU Zhi-qiang(University of Mechanical and Telecommunication Engineering,Zhengzhou Technology and Business University,Zhengzhou Henan 451400,China;School of Computer and Information Engineering,Henan University,Kaifeng Henan 475004,China)
出处
《计算机仿真》
北大核心
2018年第7期318-322,共5页
Computer Simulation
基金
国家自然科学基金项目(61305042)
河南省高等学校重点课题(158520008)
关键词
云计算
分布式
数据存储
Cloud computing
Distributed
Data storage
