Online Analytics Processing (OLAP) is a data analysis tool, aimed at _large_ and **multi-dimensional datasets**. OLAP are used for all sorts of business intelligence & analytics reports. Involves: 1. Data extraction & preparation 2. Cube creation 3. Analysis & reporting - interacting with the cube to create views (slices) from different view points >[!tldr] Like PivotTables, basically OLAP processing involves the creation of **Cube**s - these are **multidimensional cubes**, aka **hypercubes**. Data Cubes can be have derived data pre-calculated to enhance speed. They may also use calculate-on-demand setups, which are slower but take up less space. # Types ## Multidimensional OLAP Essentially just "OLAP". Involves building hypercubes, which are typically copies of the underlying datasets from which they are derived. ## Relational OLAP This allows for OLAP slicing and dicing to take place on top of traditional [[Relational Databases]]. This gives a number of advantages, but have a slower top-speed (amongst other disadvantages). ## Hybrid OLAP Basically you have a relational OLAP setup, but the option to build data cubes on demand on an as-needed basis. There are apparently a number of ways to do this. # Queries There isn't really a universal standard like there is in [[SQL]]. There's a few main-_ish_ ones. LINQ, MDX, perhaps others. # Clients - Excel can do OLAP stuff - Microsoft offers others - MicroStrategy - Oracle has some - There are some [[Open Source]] ones, but none I recognized. **** # More ## Source - Grad School ## Related