[计算机软件及应用]Chapter 2 The DW Environment.ppt

Data Homogeneity and Heterogeneity Data in the data warehouse is subdivided by: Subject area Table Occurrences of data within table Data warehouse serves as a basis for many different components * Purging warehouse data Data has its own life cycle within the warehouse 区别于一般数据库中的数据清理 Ways in which data is purged: Added to a rolling summary file where detail is lost Transferred to a bulk storage medium from a high-performance medium Purged from the system Transferred from one level of the architecture to another, such as from the operational level to the data warehouse level 2.8 * Reporting and the Architected Environment 2.9 * The Operational Window of Opportunity Operational window of data The time horizon of archival data found in the operational environment Varies from industry to industry, depends on what industry one company is in The length of the operational window is important to the DSS analyst Individual-item analysis can be done on data found within the operational window Unable to do trend analysis 2.10 * Incorrect Data in the Data Warehouse Ways to handle incorrect data in DW Go back into the data warehouse and find the offending entry The integrity of the data has been destroyed The update must be done Enter offsetting entries Many entries may have to be corrected Reset the state to the current state Application and procedural conventions are required Does not accurately account for the error that has been made 2.11 * Summary Data Warehouse definition Granularity of data Partitioning of data Data structure in the data warehouse Heterogeneous data in the data warehouse Reporting and the architected environment Data purging Incorrect data issues * Questions ? The data warehouse is defined as ? The benefit of granularity? Data structures in the data warehouse environment? * * 源数据（早期细节级数据）经过综合后，首先进入当前细节级，然后根据应用的需求，通过预运算将数据聚合成轻度综合和高度综合级。 粒度越大，表示细节程度越低，综合程度越高。 注意：轻度和高度只是相对概念，没有绝对的界限；一般DW中存在的级别一般很多。 综合后的数据也包括元数据，它描述的是提取和综合后的数据的组织方式。 * * 系统层上分区时，DBMS希