OLAP is defined as the analytical processing system that supports Business Intelligence by accessing data in various formats. Here, the cube is getting designed in between database and presentation layout. It’s synonymous to metadata where the structure is available and we query the OLAP (Online Analytical Processing) cube with specific MDX (Multi-Dimensional Expression) type of query along with extracting the data from data warehouse. It can be divided into two types:
rOLAP or relational OLAP – uses data warehouse in RDBMS. Here, we have metadata in cube and extract the relational database.
mOLAP or memory OLAP – Here, data is stored in cubes
What is Mondrian?
It is ROLAP (Relational Online Analytical Processing) server, based on Java and is in the domain of reporting and data warehousing. It is useful when we use for analysis that involves drilling down data. It’s an open-source OLAP server and supports MDX query language. The Mondrian schema is universal metadata descriptor supported by almost any OLAP client tool.
Here, we design two types of data-models in data warehouse.
Star Schema Model – It is star shaped with two tables; a fact table and a dimension table. Dimension is basically the entity. For example, in an organization, we have a master data with the customer list, product list and so on. One of the important dimension here is time dimension. The time dimension stores time in a way with elements like date, month and year. The fact table contains the foreign key to the dimension table and the measures. Measure is a numerical value useful for a business user and is in the fact table. So the dimension and the fact table link make the Star Schema Model.
Multi-Dimensional Cube – It contains time dimensions, figures and location. It’s a cube to view data where we can slice and dice data.
Snowflake Schema – Another type of schema is snowflake schema where it has the parent-child relationship. The basic structure will remain the same but the dimension will have one more table linked to it.
It is always good to use star schema because it has lesser joints. The lesser joints we have in our query, the faster the results.
Got a question for us? Mention them in the comments section and we will get back to you.