Data Warehouse, Data Mart and Business Intelligence Essay

Data Warehouse, Data Mart and Business Intelligence - Essay Example Data Warehouses, Data Marts and Databases A data warehouse refers to a data storage location used to secure, archive, and analyze data. It comprises of many integrated databases in an organization. Data stored in a data warehouse must be easily accessible to facilitate the daily operations of an organization. There are several types of data ware houses. There are offline operational data warehouses where data is copied from real time data networks and stored offline. Offline data warehouses store integrated data that is frequently updated and can be easily accessed. Real-time data warehouses are updated whenever new data comes in, for example in point of sale systems. Integrated data warehouses can be accessed by other systems (Jensen, Pedersen, and Thomsen, 2010). Data marts refer to smaller data warehouses covering a specific department or subject. They differ from data warehouses in that they are less complex, and are easier to develop and maintain. Data warehouses also focus on many subject areas and collect their data from various sources while data marts deal with one subject and collect data from few sources. There are dependent and independent data marts. Dependent data marts source their data from a functional central data warehouse while independent data marts get data from external sources. A data mart can be a small division of a data warehouse (Jensen, Pedersen, and Thomsen, 2010).... Databases contain records of data that can be easily accessed. While databases are designed to record and store data, data warehouses are designed to respond to critical business queries. All data warehouses are databases but few databases can be considered to be data warehouses. Databases are usually online transaction processing systems for recording transactions while data warehouses are online analytical processing systems for querying and analyzing data (Jensen, Pedersen, and Thomsen, 2010). Data Warehouse Architectures and Tools Data warehouses are developed using several steps including data collection, data cleansing, data aggregation, and analysis and presentation. Data collection involves identifying the suitable data for the warehouse and where it can be sourced from. In data cleansing and transformation, the collected data is restructured to make it usable for reporting, querying, and analysis. Data aggregation and analysis involves the use of query tools to transfer data from the central data warehouse and processing it to produce the required results. Presentation involves giving end results to the users in form of text, charts or tables (Barry, 2003). There are various data warehouse architectures varying from one organization to another depending on their data. These architectures include independent data marts, hub-and-spoke, federated, centralized data warehouse and data mart bus architecture that has linked dimensional data marts. Independent data marts architecture involves developing autonomous marts with different data definitions, measures and dimensions. Data bus mart with linked dimensional data marts architecture is designed to meet the needs of a specific business process. It involves the development of one