支持增量式数据仓库建设的多维数据模型

Multidimensional Data Model the Building of Incremental Data Warehouse

由于数据仓库很难做到从建模初期就完全考虑到多维分析的所有维及其层次,需要一个易于实现模式演变的数据模型及建模工具来支持增量式数据仓库的建设,因而提出了一个可以支持模式演变的数据模型,并给出其形式定义及在关系数据库上的设计与实现.该模型将数据立方体与复杂维模式分开并将度量维度化,其简单灵活的多维数据立方体结构有效地避免了因度量或维路径层次增减造成的数据立方体重建,在实际应用中取得了良好的效果,满足了复杂多样、不断变化的OLAP(on-lineanalyticalprocessing)需求.

In view of all dimensions and layers of multi-dimension analyzing at the beginning of modeling data warehouse system. A data model evolved easily, as well as corresponding modeling tools are needed for the building of incremental data warehouse. In this paper, a data model supporting model evolving is presented. The model＇s formalized definition, as well as it＇ s design and impliment on RDB, is introduced. The data model separate data cubes and complicated dimension patterns, and make the measures dimensioned. The flexible structure of multi-dimensional data cube avoid rebuild data cubes caused by the change of measures and layers efficiently. The model stetisfy the requirement of the complicated and continuously changed OLAP.