Fast Computation of Sparse Data Cubes with Constraints  

作　　者：FengYu-cai ChenChang-qing FengJian-lin XiangLong-gang 

机构地区：[1]SchoolofComputerScienceandTechnology,HuazhongUniversityofScienceandTechnology,Wuhan430074,Hubei,China

出　　处：《Wuhan University Journal of Natural Sciences》2004年第2期167-172,共6页武汉大学学报（自然科学英文版）

基　　金：SupportedbytheE GovernmentProjectoftheMinistryofScienceandTechnologyofChina (2 0 0 1BA1 1 0B0 1 )

摘　　要：For a data cube there are always constraints between dimensions or among attributes in a dimension, such as functional dependencies. We introduce the problem that when there are functional dependencies, how to use them to speed up the computation of sparse data cubes. A new algorithm CFD (Computation by Functional Dependencies) is presented to satisfy this demand. CFD determines the order of dimensions by considering cardinalities of dimensions and functional dependencies between dimensions together, thus reduce the number of partitions for such dimensions. CFD also combines partitioning from bottom to up and aggregate computation from top to bottom to speed up the computation further. CFD can efficiently compute a data cube with hierarchies in a dimension from the smallest granularity to the coarsest one. Key words sparse data cube - functional dependency - dimension - partition - CFD CLC number TP 311 Foundation item: Supported by the E-Government Project of the Ministry of Science and Technology of China (2001BA110B01)Biography: Feng Yu-cai (1945-), male, Professor, research direction: database system.For a data cube there are always constraints between dimensions or among attributes in a dimension, such as functional dependencies. We introduce the problem that when there are functional dependencies, how to use them to speed up the computation of sparse data cubes. A new algorithm CFD (Computation by Functional Dependencies) is presented to satisfy this demand. CFD determines the order of dimensions by considering cardinalities of dimensions and functional dependencies between dimensions together, thus reduce the number of partitions for such dimensions. CFD also combines partitioning from bottom to up and aggregate computation from top to bottom to speed up the computation further. CFD can efficiently compute a data cube with hierarchies in a dimension from the smallest granularity to the coarsest one. Key words sparse data cube - functional dependency - dimension - partition - CFD CLC number TP 311 Foundation item: Supported by the E-Government Project of the Ministry of Science and Technology of China (2001BA110B01)Biography: Feng Yu-cai (1945-), male, Professor, research direction: database system.

关 键 词：sparse data cube functional dependency DIMENSION PARTITION CFD 

分 类 号：TN911[电子电信—通信与信息系统]