【作者】 郭少松； 包小源； 陈薇； 王腾蛟； 雷凯； 杨冬青；

【Author】 Guo Shaosong~(1,2),Bao Xiaoyuan~2,Chen Wei~(1,2),Wang Tengjiao~(1,2),Lei Kai~3,and Yang Dongqing~(1,2) 1(Key Laboratory of High Confidence Software Technologies(Ministry of Education)) 2(School of EECS,Peking University,Beijing 100871) 3(The Shenzhen Key Lab for Cloud Computing Technology and Applications(SPCCTA),Peking University Shenzhen Graduate School)

【机构】 高可信软件技术教育部重点实验室； 北京大学信息科学技术学院； 深圳市云计算关键技术与应用重点实验室(SPCCTA),北京大学深圳研究生院；

【摘要】 含谓词的XPath查询为小枝查询(twig query),它是XQuery最基本的查询模式.小枝查询算法有两大类:小枝分解和整体小枝连接,其中TwigStack系列算法处理小枝查询的效率是最优的,但其缺点在于不支持Twig Query中含"OR"以及通配符"*"的查询;在大型商业数据库管理系统中,TurboXPath是DB2用于处理小枝查询的基本算法,它对XPath的支持较全面,但由于TurboXPath是针对XML数据流查询处理设计的,因而对数据库中Native存储的XML数据的查询处理效率较低;针对已有方法的不足,设计了一种纯XML存储方式,以DDE编码为基础,提出了TurboStack算法,并通过大量实验验证了该方法的有效性.更多还原

【Abstract】 The XPath with predicates is called Twig Query,which is the basic query model of XQuery.There are two kinds of Twig Query algorithms,namely twig split and holistic twig join. Among all the algorithms,the TwigStack algorithm in the series has the best efficiency,but it can’t support the XPath with "OR" or wildcard " * ".In the big commercial database management systems, the TurboXPath algorithm is adopted by DB2 for processing Twig Query.Its support for XPath is more comprehensive,but it is developed in view of XML stream so its efficiency in processing XML data with native storage model in database is low.In view of these disadvantages,we design a kind of native xml storage and propose TurboStack algorithm based on DDE labeling scheme.It proves that this algorithm is efficient according to a lot of experiments.更多还原