Information Technology Reference
In-Depth Information
Data Warehouses
A
data warehouse
is a database that holds business information from many sources in the
enterprise, covering all aspects of the company's processes, products, and customers. The
data warehouse provides business users with a multidimensional view of the data they need
to analyze business conditions. Data warehouses allow managers to
drill down
to get more
detail or
roll up
to take detailed data and generate aggregate or summary reports. A data
warehouse is designed specifically to support management decision making, not to meet the
needs of transaction processing systems. A data warehouse stores historical data that has been
extracted from operational systems and external data sources (see Figure 5.17). This opera-
tional and external data is “cleaned up” to remove inconsistencies and integrated to create a
new information database that is more suitable for business analysis.
data warehouse
A database that collects business
information from many sources in
the enterprise, covering all aspects
of the company's processes, prod-
ucts, and customers.
Figure 5.17
Elements of a Data Warehouse
Relational
databases
Data
extraction
process
Data
cleanup
process
Flat
files
Data
warehouse
Spreadsheets
End-user access
Query and
analysis tools
Data warehouses typically start out as very large databases, containing millions and even
hundreds of millions of data records. As this data is collected from the various production
systems, a historical database is built that business analysts can use. To keep it fresh and
accurate, the data warehouse receives regular updates. Old data that is no longer needed is
purged from the data warehouse. Updating the data warehouse must be fast, efficient, and
automated, or the ultimate value of the data warehouse is sacrificed. It is common for a data
warehouse to contain from three to ten years of current and historical data. Data-cleaning
tools can merge data from many sources into one database, automate data collection and
verification, delete unwanted data, and maintain data in a database management system.
Data warehouses can also get data from unique sources. Oracle's Warehouse Management
software, for example, can accept information from Radio Frequency Identification (RFID)
technology, which is being used to tag products as they are shipped or moved from one
location to another. Instead of recalling hundreds of thousands of cars because of a possible
defective part, automotive companies could determine exactly which cars had the defective
parts and only recall the 10,000 cars with the bad parts using RFID. The savings would
be huge.
