Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


To illustrate the effect of granularity on the ability to do queries, in Figure 2.13, the following query is made: “Did Cass Squire call his girlfriend in Boston last week?”


With a low level of granularity, the query can be answered. It may take a lot of resources to thumb through a lot of records, but at the end of the day, whether Cass called his girlfriend in Boston last week can be determined.


But with a high level of detail, there is no way to definitively answer the question. If all that is kept about Cass Squire in the data warehouse is the total number of calls that he made for a given week or month, whether one of those calls went to Boston cannot be determined.


When doing DSS processing in the data warehouse environment, examining only a single event is rare. A collective view of data is much more common. To achieve a collective view requires looking at a large number of records.


For example, suppose the following collective query is made: “On the average, how many long-distance phone calls did people from Washington make last month?”


granularity


low level of detail


high level of detail




EXAMPLE:


the details of every phone call made by a customer for a month



EXAMPLE:


the summary of phone calls made by a customer for a month



“Did Cass Squire call his girlfriend in Boston last week?”


• Can be answered, even though    • Cannot be answered in any case

Скачать в pdf «Building the Data Warehouse»