分布式数据库系统的数据分布策略研究  

Research on Data Distribution Strategy of Distributed Database System

作　　者：包全磊[1] BAO Quanei(Taiyuan University,Taiyuan Shanxi 030001,China)

机构地区：[1]太原学院,山西太原030001

出　　处：《信息与电脑》2022年第24期196-198,共3页Information & Computer

摘　　要：为降低局域网数据通信成本、提高整体数据分布与处理性能,文章采取分布式数据库的任务处理方案。首先,将数据处理任务分配到各个服务器站点,按照统计数据的完整性、重构性和不相交性要求,利用分簇-遗传算法将整个站点集合按照“通信代价阈值”设定簇集;其次,以“簇”为对象进行数据分片和副本分配,将重要数据分片到访问频率较高的站点;最后,经由传输控制协议/网际协议(Transmission Control Protocol/Internet Protocol,TCP/IP)、ZigBee网络协议等互联网络,形成数据库系统逻辑上的统一管理,可保证系统数据分布的效率和质量。In order to reduce the data communication cost of the Local Area Network and improve the overall data distribution and processing performance, this paper adopts the task processing scheme of the distributed database. Firstly, assigns the data processing tasks to each server site, and uses clustering genetic algorithm to set the cluster of the entire site set according to the “communication cost threshold”according to the integrity, reconfiguration, and disjoint requirements of statistical data. Secondly, the “cluster” is taken as the object for data sharding and replica allocation. The important data is sharded to sites with high access frequency. Finally, through the Transmission Control Protocol/Internet Protocol(TCP/IP) network protocol, ZigBee network protocol and other interconnection networks, a unified management of database system logic is formed, which can ensure the efficiency and quality of system data distribution.

关 键 词：分布式数据库系统 数据分布 策略 遗传算法 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]