Online Analytical Processing (OLAP) is a category of software for performing multidimensional analysis at high speeds on large volumes of business data from a data warehouse or centralized data store. It provides capabilities for complex calculations, trend analysis, and sophisticated data modeling. OLAP is part of a broader category of business intelligence, which includes relational databases, report writing, and data mining.
OLAP serves as the foundation for many business applications such as business process management, planning, budgeting, forecasting, financial reporting, knowledge discovery, and simulation models. Users of OLAP can perform ad hoc analysis of data in multiple dimensions, consequently providing the insight and understanding needed for decision making.
OLAP cube
At the core of the OLAP concept is an OLAP cube, a data structure optimized for quick data analysis. It consists of numeric facts known as measures that are categorized by dimensions. Cubes enable four basic types of multidimensional data analysis:
- Drill down: Data is fragmented into smaller parts. It converts less-detailed data into more-detailed data by either moving down in the concept hierarchy or adding a new dimension to the cube.
- Roll up: The opposite of the drill down process. Data is aggregated by moving up in the concept hierarchy or by reducing the number of dimensions.
- Slice and dice: For slice, a new sub-cube is created by selecting a single dimension from the main cube. For dice, a sub-cube is isolated by selecting several main cube dimensions.
- Pivot (rotate): The cube is rotated to display a new representation of data, which enables the dynamic, multidimensional views of data.
Types of OLAP
There are multiple types of OLAP, but three main ones exist:
- Relational OLAP (ROLAP): An extension of relational database management systems. It is multidimensional data analysis that operates directly on data on relational tables without reorganizing the data into a cube first.
- Multidimensional OLAP (MOLAP): The fastest and most practical type of multidimensional data analysis. It works directly with a multidimensional OLAP cube.
- Hybrid OLAP (HOLAP): Aggregated totals are stored in a multidimensional database while the detailed data is stored in the relational database. Offers both the efficiency of the ROLAP model with the performance of the MOLAP model.
Other types of OLAP include Desktop OLAP (DOLAP), Web OLAP (WOLAP), Mobile OLAP, and Spatial OLAP.
