Hybrid Storage (HOLAP)

Roll-up and drill-down, slice and dice, drill-across, pivoting, and rotation.

To jump to a higher level of aggregation.

To view data across a single plane.

ROLAP stores data in two-dimensional tables, while MOLAP stores data in a multidimensional database management system.

SCD type 1, SCD type 2, and SCD type 3.

The corresponding data record is overwritten, resulting in the loss of previous information.

A validity interval is added to each data record, allowing for full historicization of dimensional changes.

The existing data record is extended with a new attribute containing the new designation.

The study goals for this unit include learning basic modeling techniques, understanding the analysis possibilities offered by OLAP cubes, knowing how multidimensional models are physically stored, and being aware of the options available for historicizing dimensions.

A multidimensional data model is more flexible than a relational data model because it allows data to be located across multiple dimensions, providing flexible views of the data. In contrast, a relational data model locates data in two dimensions.

Star and snowflake schemas are used for performance-oriented modeling of multidimensional spaces in business intelligence.

The purpose of slowly changing dimensions (SCD) in a data warehouse is to handle changes in dimension tables over time.

The three different ways of recording changes in dimension tables for slowly changing dimensions (SCD) are SCD type 2, SCD type 3, and SCD type 4.

The two different database structures for the physical implementation of a multidimensional data model are the star schema and the snowflake schema.

The purpose of the entity relationship model (ERM) is to structure the communication between users and developers in the conceptual phase of application development, as well as to serve as a basis for database design in the implementation phase.

The major disadvantage of not using a model like ERM when creating relational databases is the potential storage of redundant information, which compromises the consistency of the database and can lead to anomalies.

Normalization is a procedure used to avoid redundancies and inconsistencies in relational databases. It transforms databases into data structures that are free of redundancy or minimize redundancy. It helps manage data more efficiently.

An OLAP cube is a multidimensional data space in the online analytical processing (OLAP) cube model. It consists of dimensions (edges of the cube) and key figures or facts (values at the coordinates inside the cube). Dimensions describe key business figures, while facts are operation-related, quantitative variables.

HOLAP stands for Hybrid Online Analytical Processing. It combines the strengths of relational and multidimensional concepts by using both a relational and a multidimensional database for storage. Detailed data are stored in the relational database, while aggregated data are stored in the multidimensional database. Data access is achieved through a multidimensional query tool.

The two different database structures used in the physical implementation of a multidimensional data model are star schema and snowflake schema.

The star schema consists of a fact table in which the keys for a data cube are managed. The dimension tables are arranged in a star shape around the fact table. There is a separate table for each dimension. Relationships only exist between the fact table and the dimension table, not between the dimension tables. The link between the fact table and the dimension table is achieved by including the primary key of the fact table as a foreign key in the primary keys of the dimension tables.

The star schema has dimension tables arranged in a star shape around the fact table, with relationships only between the fact table and the dimension tables. The snowflake schema has dimension tables surrounded by dimension tables, resulting in hierarchies being split into additional dimension tables. The snowflake schema is normalized with regard to functional dependencies, which can result in reduced performance for queries compared to the star schema.