一种支持复杂查询的多维云数据管理索引机制

Multi-dimensional Indexing for Complex Query in Cloud Data Management

针对当前分布式云数据管理系统的数据索引不支持复杂查询的问题,提出一种支持复杂查询的多维云数据索引机制PR-Chord.PR-Chord由基于PR四叉树的全局索引PR-Index和覆盖网络路由协议Chord组成.将由多维数据的值域所形成的多维空间根据PR四叉树的剖分规则平均划分为若干个超矩形,利用改进的PR四叉树技术建立树形索引PR-Index,分层索引这些超矩形空间.多维数据的复杂查询转化为对PR-Index叶节点的查询.设计了数据查询、插入和删除算法,以支持多维查询和区间查询.PR-Index索引本身不存储多维数据,因此维护代价为零.PR-Chord具有负载均衡、算法简单的优点.实验证明PRChord具有良好的查询效率.

To solve themultidimensional query and range query of some cloud data managements,we propose a multidimensional indexing mechanism named PRChord.PR-Chord consists of a global index named PR-Index and the overlay network routing protocol named Chord.PR-Chord divides the multidimensional space formed by the range of multidimensional data into a number of hyper rectangles equally.Based on the improved structure of PR quadtree,PR-Index is the hierarchical index of these hyper rectangular spaces.The complex query of multi-dimensional data is transformed into the query of leaf nodes of PRIndex.Algorithms of data query,insertion and deletionare designed to solve the multidimension query and range query.PR-Index index itself does not store multidimensional data,so the maintenance cost is zero.PR-Chord has the advantages of load balancing and simple algorithm.Experiments show that PRChord has good query efficiency.