【作者】 张新； 孟小峰； 朱金清； 王伟； 黄静；

【Author】 Zhang Xin,Meng Xiaofeng,Zhu Jinqing,Wang Wei,and Huang Jing (School of Information,Renmin University of China,Beijing 100872)

【机构】 中国人民大学信息学院；

【摘要】 XML数据在数据库中的存储模式对XML数据的查询、索引及更新有重要的影响.而目前许多XML存储方法在更新上都需要较高的代价.提出一种Native XML存储方法OrientStore^?,可以完全保留XML树结构信息,同时还具有如下特点:1)易于对XML数据建立各种索引;2)存储记录间相互独立,进行更新时,可以减少对XML存储及索引的修改,减小了更新的代价;3)在Native XML数据库系统OrientX中实现了这种存储模式.另外,在这种存储模式基础上提出一种基于空间利用率的XML存储更新算法.并通过实验比较了在不同存储方法上的查询与更新效率.更多还原

【Abstract】 The storage strategy for XML data in database will affect the efficiency of querying,indexing and updating significantly.The cost is high for most of state-of-the-art XML storage strategies when considering update.In this paper,an optimized native XML storage strategy,called OrientStore~+,is proposed,which can keep the structure relationship between nodes completely and the new storage can also satisfy the following virtues:1) it can facilitate building index on XML data;2) the records in the storage are independent,which can reduce the modification of the storage data and index after update,reducing the update cost;and 3) the storage strategy is implemented in native XML database system OrientX.In this paper,a new update algorithm which is based on the space usage ratio,is proposed.Finally,the efficiency of query and update using different storage strategies is compared through extensive experiments.更多还原