Building the Data Warehouse

Скачать в pdf «Building the Data Warehouse»


There is significantly less volume of data in the lightly summarized database than there is in the detailed database. Of course, there is a limit to the level of detail that can be accessed in the lightly summarized database.


id d


lightly summarized data


d dl

30 days’ detail    lightly summarized


J Jones


April 12 6:01 pm to 6:12 pm 415-566-9982 operator assisted April 12 6:15 pm to 6:16 pm 415-334-8847 long distance April 12 6:23 pm to 6:38 pm 408-223-7745


April 13 9:12 am to 9:23 am 408-223-7745


April 13 10:15 am to 10:21 am 408-223-7745 operator assisted April 15 11:01 am to 11:21 am 415-964-4738


April 15 11:39 am to 12:01 pm 703-570-5770 incomplete April 15 12:10 pm to 12:46 pm 703-841-5770 wrong number April 16 12:34 pm to 12:56 pm 415-964-3130



April J Jones


number of calls — 45 avg length of call — 14 minutes number of long distance calls — 18 number of operator assisted calls — 2 number of incomplete calls — 1



number of bytes required to house a record—225


For a single customer for a month, an average of 45,000 bytes are required to house 200 records.


Figure 2.16 With light summarization data, large quantities of data can be represented compactly.


The second tier of data in the data warehouse—the lowest level of granularity—is stored in the true archival level of data, as shown in Figure 2.17.


At the true archival level of data, all the detail coming from the operational environment is stored. There is truly a multitude of data at this level. For that reason, it makes sense to store the data on a medium such as magnetic tape or another bulk storage medium because the volume of data is so large.

Скачать в pdf «Building the Data Warehouse»