一种基于HBase的语义数据存储模型  

A SEMANTIC DATA STORAGE MODEL BASED ON HBASE

作　　者：翟社平 高山 郭琳 李兆兆 Zhai Sheping;Gao Sha;Guo Lin ;Li Zhaozhao(School of Computer Science,Xi’an University of Posts and Telecommunications,Xi’an 710121,Shaanxi,China)

机构地区：[1]西安邮电大学计算机学院,陕西西安710121

出　　处：《计算机应用与软件》2018年第3期15-20,共6页Computer Applications and Software

基　　金：陕西省自然科学基金项目(2012JM8044);陕西省社会科学基金项目(2016N008);陕西省教育厅科学研究计划项目(12JK0733);西安市社会科学规划基金项目(17X63);西安邮电大学研究生创新基金项目(114-602080106)

摘　　要：随着语义Web的快速发展与应用,如何高效地组织管理大规模语义数据成为语义Web研究中的一个关键问题。为了高效地执行各种查询操作,在充分利用语义数据查询特征的基础上,提出一种基于分布式数据库HBase的资源描述框架RDF(resource description framework)数据存储模型。该模型将以谓语为索引的RDF数据存放在P_OS和P_SO两类索引表中,同时给出改进后的查询索引策略。在数据加载方面,通过自定义Bulk Loading的方式快速将数据上传至HBase存储表中。理论分析和实验结果表明,提出的存储模型能够快速响应基于谓语约束的查询服务;另一方面,Bulk loading并行数据加载方法具有较高的加速比,能够显著缩短数据加载时间。With the rapid development and application of Semantic Web,how to organize and manage large-scale semantic data becomes a key problem in Semantic Web research.In order to perform various query operations efficiently,an RDF data storage model based on distributed database HBase was proposed based on the semantic data query feature.The RDF data with predicate index were stored in P_OS and P_SO index tables,and the improved query indexing strategy was given.In terms of data loading,the data was uploaded to the HBase storage table quickly by customizing Bulk Loading.The theoretical analysis and experimental results showed that the storage model proposed in this paper could respond quickly to the query service based on predicate constraints.On the other hand,the loading method of Bulk loading parallel data had a higher speedup,which could significantly shorten the data load time.

关 键 词：HBASE 资源描述框架 语义网 语义数据存储中图分类号 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]