This paper proposed a new structure of data cube. This cube structure changed data storage mode of dwarf cube, could keep the basic roll-up and drill-down semantic cube, removed the prefix redundancy and suffix redundancy as far as pos sible. This structure could save storage space, ensured a clear cubic lattice structure, and possessed higher storage efficiency and less query response time than dwarf cube. For point queries and range queries, it could return query results very quickly. This data cube structure scaled well on the condition of sparse cube with large amount of data. It proposed the construction algorithm, search algorithm and incremental update algorithm of the new data cube, and also studied the corresponding iceberg cube.%在侏儒立方体研究的基础上,提出了一种新的能够保持语义的立方体结构.这种结构改变了侏儒立方体对聚集数据的存储方式,在保持基本立方体上卷、下钻语义的前提下,尽量地去除前缀冗余、后缀冗余,节约存储空间,保证立方体清晰的结构,并且拥有比侏儒立方体更高的存储效率和查询响应速度,对点查询和范围查询能够快速地返回结果,对大数据量情况下的稀疏立方体具有良好的支持.
展开▼
