基于RDF的云制造数据在HBase中管理方式的研究.docVIP

  基于 RDF 的云制造数据在 HBase 中管理方 式的研究 曹佳硕，张宝鹏** 5 10 15 20 25 30 35 40 （北京交通大学计算机与信息技术学院网络管理研究中心，北京 100044） 摘要：随着语义网技术的不断发展与成熟，资源描述框架 RDF 被应用于越来越多的领域中， 然而随着全球全面进入信息化，数据爆炸式的增长，传统的 RDF 存储系统已无法满足用户 的数据管理要求，因此，国内外研究者逐渐将目光转向使用分布式数据库来进行 RDF 数据 的管理。本文首先对 RDF 所描述数据在 HBase 中的存储方案进行研究，对当前国际较为主 流的存储方案进行改进，使其可以在处理 RDF 可能出现的多值问题时具有更高的效率。然 后，本文对 RDF 查询语言——SPARQL 与 HBase 之间的查询接口进行了研究，在查询接口 中，结合当前的国际主流的查询逻辑，加入了语义推理部分，使传统的信息的查询由基于关 键字转变为基于关键字与基于语义相结合的查询。再次，本文在整体查询接口上引入了索引 机制与并行查询机制。最后，本文通过对比试验，对不使用索引与并行机制的查询方案与使 用索引与并行机制的查询方案进行了对比。并对实验结果进行分析，最后得出结论。 关键词：RDF；Hbase；SPARQL 中图分类号：TP301 Research of the management of the RDF-based cloud manufacturing data in Hbase CAO Jiashuo, ZHANG Baopeng (School of computer and information technology, Beijing Jiaotong University, Beijing 100044) Abstract: With the Semantic Web technology continues to evolve and mature, RDF is applied to a growing number of many fields, however, with the world enter to the information technology, the amount of the data growth fast, The traditional RDF storage system has been unable to meet the data management requirements of the user. This article first study the data described by RDF storage solutions in the HBase, and improve more mainstream storage solutions on the current international, to make it more efficient in dealing with RDF data. Then, this article discusses query interface between the RDF query language – SPARQL and HBase. In Query interface, combined with the current international mainstream query logic, adding a semantic reasoning part. Make the traditional query information based on keywords into semantic-based queries. Once again, this article introduces the indexing mechanism and parallel query mechanism overall query interface. Finally, this article conducted a comparative test, compared query program that does not use an index and parallel mechanism with using the query index and parallel mechanism. The experimental results were analyzed, and finally concluded. Keywords: RDF; Hbase; SPARQL 0 引言 “云制造”