基于Hadoop的海量交通数据混合存储查询系统设计与实现  

Design and implementation of a hybrid storage and query system based on Hadoop for massive traffic data

作　　者：余超[1] 王坚[1] 凌卫青 

机构地区：[1]同济大学CIMS研究中心,上海201804

出　　处：《信息技术与信息化》2016年第1期82-86,共5页Information Technology and Informatization

基　　金：国家自然科技基金面上项目 名称:基于群智能的突发事件区域人群疏散模型研究(71273188)

摘　　要：随着大数据时代的来临,交通行业也迎来了巨大的机遇与挑战,传统的基于关系型数据库的交通数据存储系统已无法满足应用对数据查询快速响应的需求。论文提出了一种将交通静态数据和动态数据分离的混合存储查询方案,分别使用HBase来存储RDF静态数据,Hive来存储交通动态数据。论文首先介绍了系统的设计理念和系统的设计方案,然后介绍了系统的结构设计和功能设计,最后展示了系统的原型实现。with the advent of the era of big data, the transportation industry has also ushered in a huge opportunity and challenges, traditional traffic data storage system based on relational database has been unable to meet the requirement of application for data query response. In this paper, we propose a hybrid storage and query schema that separates traffic static data and dynamic data, using HBase to store RDF static data, and Hive to store traffic dynamic data. First of all, the paper introduces the design concept and system design, and then introduces the structure design and function design of the system, and finally shows the prototype of the system.

关 键 词：RDF存储 HBASE Hive 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]