Publisher Summary Online analytical processing (OLAP) uses a snapshot of a database taken at one point in time and then puts the data into a dimensional model. The purpose of this model is to run queries that deal with aggregations of data rather than individual transactions. In traditional file systems, one can use indexes, hashing, and other tricks for the same purpose. One such structure is the cube (or hypercube); the OLAP cube is created from a star schema of tables. At the center is the fact table, which lists the core facts that make up the query. Basically, a star schema has a fact table that models the cells of a sparse array by linking them to dimension tables. Furthermore, the chapter discusses special features, which were added to make the OLAP engines practical. Treatment of nonnormalized data: this means one can load data from non-RDBMS sources. Relational OLAP (ROLAP) was developed after MOLAP. The main difference is that ROLAP does not do precomputation or store summary data in the database. ROLAP tools create dynamic SQL queries when the user requests the data.