MATLAB Interface for Databricks

The MATLAB Interface for Databricks^® is a support package designed to integrate MATLAB^® and Simulink^® with the Databricks cloud-based data platform. MATLAB and Simulink are used in engineering to design and develop complex systems, while Databricks provides tools for data engineering, machine learning, and collaborative data science through Spark™-based workloads.

Since its introduction in 2020, this interface has helped engineers connect to, analyze, and manipulate data in Databricks. Beyond individual productivity, it supports collaboration across roles by providing a shared framework for working with data. This integration enables professionals to streamline workflows, extract insights, and optimize data-driven solutions.

The MATLAB Interface for Databricks supports three key data workflows:

1. Interactive Exploration and Prototyping
   This workflow focuses on quick insights and interactive data exploration. Engineers can connect to Databricks from their local machine using Database Toolbox™ and SQL commands—without directly interacting with Spark. It is suited for small data sets and early-stage concept development, addressing the question:
   “The data I need is in Databricks—now what?”

2. Spark-Based Data Manipulation
   The second workflow involves direct interaction with Spark for scaling data analysis and transformation. Engineers can author Spark commands in MATLAB and execute them on Databricks clusters, allowing the handling of large data sets. This workflow supports iterative development while advancing algorithms and models toward production.

3. Production Deployment of MATLAB and Simulink Models
   The third workflow is designed for deploying stable MATLAB code or Simulink models in Databricks production environments. Using the MATLAB Compiler™ or Simulink Compiler™ toolboxes, engineers can move their computations closer to the data, supporting scalability and efficiency for production workloads.

By connecting MATLAB and Databricks, this interface provides flexibility to develop insights, refine models, and deploy solutions efficiently. Whether working with large data sets or production environments, the MATLAB Interface for Databricks helps users leverage both platforms for engineering and data science work.