Databricks Runtime 11.3 LTS is a long-term support version released in October 2022, powered by Apache Spark 3.3.0. It is designed for stability and reliability, receiving regular maintenance updates and security patches.

Key Features and Improvements:

• Enhanced Kinesis Connector: This connector supports reading from Kinesis Data Streams in Enhanced Fan-Out (EFO) mode, which improves throughput and dedicates resources.
• Apache Spark 3.3.0: Includes all Spark fixes and improvements from Databricks Runtime 11.2, along with additional bug fixes and optimizations.
• Machine Learning (ML) Enhancements: The Databricks Runtime 11.3 LTS for Machine Learning includes an upgraded version of sparkdl. boost with bug fixes and many popular ML packages like hyperopic, feature_store, and autonomy.

Installation:

Databricks Runtime 11.3 LTS can be installed through the Databricks Runtime Manager in your Databricks workspace. Detailed instructions can be found in the Databricks documentation.