[ad_1]
Data warehouses store large amounts of data and have two main design types: top-down and bottom-up. Data marts are collections of data based on a single concept and are managed differently in each design. Top-down is comprehensive but expensive, while bottom-up is cheaper but less complete. Both have strengths and weaknesses.
Data warehouses store large amounts of data for use in many different fields. There are two main types of data warehouse design: top-down and bottom-up. The two models have their advantages and disadvantages. Bottom-up is easier and cheaper to implement, but is less comprehensive and data correlations are more sporadic. In a top-down design, the connections between data are obvious and well established, but the data may be out of date and the system is expensive to implement.
Data marts are the central figure in data warehouse design. A data mart is a collection of data based on a single concept. Each data mart is a unique and complete subset of data. Each of these collections is fully related internally and often has connections to external data marts.
The way data marts are managed is the main difference between the two data warehouse design styles. In top-down design, data marts occur naturally as data enters the system. In bottom-up design, data marts are built directly and linked together to form the warehouse. While this may seem like a small difference, it creates a very different design.
The top-down method was the original data warehouse design. Using this method, all information held by the organization is entered into the system. Each broad topic will have its own general area within the databases. As the data is used, connections between related data points will appear and data marts will appear. Furthermore, all data in the system remains there forever, even if the data is replaced or trivialized by later information, it will remain in the system as a record of past events.
The bottom-up method of data warehouse design works in the opposite direction. A company enters information as a standalone data mart. Over time, other datasets are added to the system, either as a data mart or as part of an already existing one. When two data marts are considered sufficiently connected, they merge into one unit.
The two data warehouse designs each have their own strengths and weaknesses. The top-down method is a huge project for even smaller datasets. Since large projects are also more expensive, it is the most expensive in terms of money and manpower. If the data warehouse is finished and maintained, it’s a large collection, containing everything the business knows.
The bottom-up process is much faster and cheaper, but since data is entered as needed, the database will never actually be complete. Furthermore, the correlations between data marts are only as strong as their usage makes them. If there is a strong correlation, but no users see it, they log out.
