The definition of a data warehouse can be given literally by explaining the two words that constitute the term - data and warehouse.
Data is basically facts and information about something and by warehouse we mean a facility or location where things are stored. A more accurate definition, though, should be given for data warehousing.

Data warehousing nowadays is an architected, periodic and coordinated process of copying from numerous sources (inside as well as outisde the company) into an optimized environment capable of analytical and informational processing. What should be noted is the two keys to this definition - the fact that data is copied in a controlled way and that it is copied periodically. This is ensured by implementing the data integration processed called ETL (extraction, transformation, loading).

    We talk about a data warehouse system when it has the following qualities:
  • it is contained in the environment that is well-managed
  • it is capable of providing centralization to corporate data assets
  • it is built on a scalable and open architecture with the ability to handle future data expansion
  • it has repeatable and consistent processes that are defined for loading the data from various corporate applications
  • it provides tools to its users for an effective process of the data into information without any requirements of high-level technical support.

Data in the data warehouses is captured to allow business leaders to make well-informed decisions that are based on previous (working and not working) business data. It is generally known that to gain and sustain advantage in the economy of the present day is to leverage information better than others do. Thanks to data waresouses, platforms to deliver, manage and implement data assets are provided. So data warehousing is creating an architected solution (information-management) to enable informational and analytical processing in spite of numerous barriers that may be encountered.

Data warehousing is linked with Business Intelligence. They constitute a powerful weapon that enables businesses to go far beyond the traditional operational data organization. The challenge for any enterprise is to put all necessary data in one common format and place and data warehouses are the right architecture to meet this challenge.

Data warehouses differ form standard data management systems. They aggregate information about an area of single subject and this resource is then used by management in one of two ways: to query to gain insights on a subject or to create reports on a certain aspect. Both are read_only as no data must be deleted from a warehouse.

Data warehousing in undeniably the key technology to enable BI but it does not mean that data warehousing is impossible without BI. In fact, many organizations create Operational Data Stores (ODSs) by utilizing data warehousing technologies.

Data warehousing technology works in the following sequence:

  • bringing in data from different external sources
  • tranforming the data into one format
  • cleansing of the gathered data
  • correlating of a given work with operational information throughout the enterprise
  • reporting daily to show this correlation.

After transforming the data into a common format and loading them into a data warehouse, the next step is to create business insights - and this is the very own domain of Business Intelligence.