A data warehouse (DW) is a central repository used to store data taken from a wide range of sources. Considered a main component of modern business intelligence, data warehouses store data that is used mainly for reporting and analysis that is critical for informed business decision making. The source data for a data warehouse can come from many different places, including internal systems, third-party applications, and data syndicators.
Data warehousing is distinct from data integration and from databases used for online transaction processing (OLTP). While the three terms and processes overlap, they have very different functions. DWs store large quantities of historical data from different sources enable complex queries across all the data, while data integration is the process of combining data from various sources with differing formats. Standard databases are designed to rigorously maintain current data and usually update in real-time. On the other hand, data warehouses store both historical data as well as current information, allowing for a long-range view of vast amounts of data over time.
So, what is a data warehouse? It is a valuable store of information that is historical, up-to-date, consistent in format, uniform, and well-suited to the deep data analysis conducted by enterprises to spot trends and data patterns in the vast amount of data available to modern data processing systems.
In order for businesses to make accurate and informed decisions, they must be able to store and work with vast amounts of historical data that comes from various sources. To be most useful, the data must be fresh and granular.
For example, a database might provide the current address of a customer, whereas a data warehouse may provide all previous addresses of the customer–and thus allow the business to deduce behavior patterns that could help with marketing, sales, and more.