A data warehouse is a repository of a business organization's historical data. It is a large part of an enterprise data management system which consists of several servers running on different kinds of platforms and database management systems.
It is generally practiced that in an enterprise data management system, it is the data warehouse house which contains static data while it is the operational data store that contains dynamic data that gets frequently updated during the course of business operations.
To illustrate this further, it important to know that in an enterprise data management system environment, there may plenty of servers and database systems which constitute various data stores and these servers may be of varying platforms and database management systems come different vendors.
Each data store gather data based on the departments they are server or on other special function that they are designed to do. But during the entire business operation, these servers send their data to the operational data store which acts as the unifying areas were disparate data from various data stores are extracted and transformed into a unified structure based on the enterprise data architecture.
The process of unifying disparate data is referred to as ETL which stands for extract, transform and load. The extract and transform are mostly done in the operational data store before the transformed data is "loaded" into the data warehouse. With this picture wherein the data warehouse only get the loading part, many people get the impression that the data warehouse indeed is a mere static repository does not do a lot of things except accept data for storage.
In fact, the concept of data warehouse has been taken from the analogy with real life warehouses where good are put before the need arise to get them. And so with data, the operational data store goes to the data warehouse to get the data and process them at the operational data store area. Hence the term operational because it refers to the data currently being operated on or manipulated with.
But modern data warehouses are no longer as static as they seem or look. Data warehouses today are already managed by software application tools that have the functionality that allows the data warehouse itself to track data and perform all sorts of analysis related to the movement of data from the warehouse to the other data stores and back.
Many data warehouse employ a technology known as Online Analytical Processing (OLAP) which helps in providing answers to various multidimensional analytical queries. Most areas of business including business reporting for sales, marketing, management reporting, business process management (BPM), budgeting and forecasting, financial reporting use OALP for retrieving information from the data warehouse so that the company can spot trends and patterns as basis for the corporate decisions.
There are many companies specifically offering data warehousing software solutions which come with sophisticated proprietary intuitive functions. Many of these vendors even offer integrated solutions that add data warehousing functions with such complex features as data transformation, management, analytics and delivery components.
Having an intuitive data warehouse greatly increases overall performance of the enterprise data management system because the data warehouse can already share some of the load which is supposed to be for the operational data stores which tackles very labor intensive processes from the on-going business operations.