Data Relationships, Data Modelling Schemas, and Joins in Power BI.

Data modelling in Power BI involves defining relationships between tables to store, relate, and query data effectively. The core responsibilities of a data model include establishing relationships between tables, enabling effective analysis, improving performance and efficiency, and supporting reusability and maintainability.

A well-designed data model impacts DAX calculations by simplifying and making them more reliable in a star schema. It also improves performance by optimising relationships for the storage and query engines, enhances scalability, and reduces the risk of modelling errors for better maintainability.

Three data modelling approaches include flat tables, star schema, and snowflake schema.

Flat table definition stores all data in a single table, making it the simplest structure but causing severe data redundancy, performance problems, maintenance challenges, and complicated DAX calculations. It's recommended for extremely small and simple datasets with low performance and scalability requirements.

Star schema, a mature dimensional modelling approach, classifies tables into dimension tables (describing business entities) and fact tables (storing measurable business events). It provides optimal performance, simplified DAX, easy understanding, high scalability, and efficient compression. The star schema is recommended for Power BI production models, supporting multiple business subjects, reusable dimensions, large data volumes, and a good DAX experience.

Snowflake schema is an extension of the star schema where dimension tables are further normalised into multiple related tables. This reduces data redundancy, suits already-normalised source systems, and allows for more flexible hierarchy management. However, it introduces reduced data redundancy, a more complex structure requiring additional management, and potential performance impacts due to extra relationships.