In computing parlance, data warehouse concepts are methods, styles, and principles used in databases that are employed for reporting and subsequent data analyses. A data warehouse may be further subdivided into what are known as data marts that serve as storage of data subsets. Stored data in the data warehouse are composed of integrated information from day-to-day company operations such as procurement, sales, and marketing, among other branches that accumulate, store, and use data. Specific groups of data may go through an operational data store where additional operations are conducted before proceeding to be utilized in the data warehouse for purposes of reporting.
Data warehouse concepts are mainly focused on the capacity for data storage. The main data source is sanitized, converted, and eventually catalogued for future use of company decision-makers and other business authorities. Utilization concepts include mining for useful data, analytical processing, and corporate decision-making. Operational concepts include data retrieval, data analysis, data extraction, data conversion, data loading, and management of the data dictionary. All these various concepts come into play for a successfully operational and beneficial data warehousing system.
Data warehouse concepts were pioneered in the late 1980s by Barry Devlin and Paul Murphy, both IBM researchers. The basic concept was to create a data architecture model that can manage data flows from operational systems to decision support environments where much of business intelligence is produced. The model greatly reduced data maintenance costs for companies and institutions. It also greatly addressed the old problems of data redundancy that confused many end user groups of data. While several decision-making groups use the same stored data, varying processes serve several specific needs.
The top-down data warehouse design uses and stores data at its finest detail. This is done with the use of dimensional data marts that house requisite data needed by specific departments for specific business processes. This makes the data warehouse the central station of the corporate information factory. From this center, business intelligence is produced and business management is applied. This is one of the most useful and widely used data warehouse concepts today.
Cutting-edge data warehouse concepts now also include other business intelligence tools used to retrieve, convert, and stock data, as well as tools to extract and manage metadata. Whether the concepts are traditional or state-of-the art, the benefits of a data warehouse have largely remained the same. All data warehouse concepts serve a data architecture that secures and processes information from several source transaction systems.