基于时间维层次查询频率的数据仓库粒度模型

数据仓库中的事实数据一般以最小粒度存储。而大量的细粒度数据具有很大的随机性,很少直接进行分析和处理,往往被聚集到一定层次的粗粒度数据。另一方面若采用ROLAP存储数据,则大量的细粒度数据将会影响查询的效率。本文介绍了一种基于时间维层次查询频率的粒度调整模型,它能根据用户在时间维层次的查询频率实现对数据粒度的调整。

Granularity Model of Data Warehouse Based on the Time Dimension Hierarchy Query Frequency

The fact data in the data warehouse is generally stored in the minimum granularity.But A large number of fine-granularity data have great randomness, is seldom used to analysis and process directly.We are often gathered the data to a certain level of coarse-granularity data. On the other hand, if the data stored by the ROLAP, a lot of fine-granularity data will affect the efficiency of the query. This article describes a granularity adjustment model based on the time dimension hierarchy query frequency.This model can adjust the data granularity by the time dimension hierarchy query frequency.