Appendix BC. Data Warehousing Concepts

To facilitate data analysis and to eliminate redundancies, the use of a data warehouse may be desirable. A data warehouse is usually a single database that houses data relating to the organization's business functions. Most data warehouses are housed in a client-server database system like Microsoft SQL Server. Client-server database applications are usually very large and have many concurrent users. Relational databases like Microsoft Access have size and user limits that are easily exceeded as an organization's data needs grow. Smaller organizations can use Microsoft Access or Microsoft SQL Server Express and if their data requirements grow to a level that requires more capacity; the existing system can easily scale to a larger more robust commercial version of SQL Server.

The goal of creating a data warehouse is usually to have a single reliable source of data that is relatively quick and easy to access. The data structure in a warehouse is usually denormalized and includes data that has been pre-summarized. The data structure is usually easier for analysts to understand and use.

Most organizations have multiple data sources that are not compatible in everyday use. The loading process in a warehouse will incorporate the data from all these sources to store it in a consistent form. There are several ways to import the data depending on whether you can connect directly with the original source using an ODBC or other direct connection or if you ...

Get Microsoft® Access® Small Business Solutions: State-of-the-Art Database Models for Sales, Marketing, Customer Management, and More Key Business Activities now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.