Data Modeling is a key aspect of data management for any large data warehousing project as it creates the blueprint for implementing a data warehouse for the data analysts, ETL and BI teams to follow. Given the explosion of data stored and processed, the speed and performance of data processing and access queries are heavily dependent on how the data is modeled both logically and physically. Physical modeling for Hadoop must also take into account the multiple storages and query engines in the Hadoop ecosystem to select from. In this article, we cover the best practices for data modeling in Hadoop.