Automatic Change Data Feed is Generally Available on Databricks
Databricks has announced the general availability of Automatic Change Data Feed (Auto CDF). This feature computes row-level changes at query time using row tracking, eliminating the need to manually enable change data feed on Delta Lake or Apache Iceberg v3 tables. Auto CDF removes write-time overhead, improving MERGE and UPDATE operations by approximately 15% when querying for changes. It supports batch queries, Structured Streaming, and Delta Lake Sharing, requiring Databricks Runtime 19 LTS or higher with row tracking enabled.
Features (1) ›
- Automatic Change Data Feed Reaches General Availability
The Automatic Change Data Feed (Auto CDF) now computes row-level changes at query time using row tracking, removing the need to manually enable CDF on Delta Lake or Apache Iceberg v3 tables. This feature removes write-time overhead and accelerates MERGE and UPDATE operations by about 15%. It requires Databricks Runtime 19 LTS or above with row tracking enabled.
https://docs.databricks.com/aws/en/release-notes/product/2026/september#automatic-change-data-feed-is-now-generally-available
Related releases
- Databricks Runtime 19 introduces Automatic Change Data Feed GA for Delta Lake Databricks Release Notes ·
- Databricks Delta Lake Sharing with Iceberg Reads Reaches GA Databricks Release Notes ·
- OpenAI GPT-6 Astra Now Available on Unity Gateway Databricks Release Notes ·
- Databricks Adds %uv Commands for Python Project Environment Management in Serverless Notebooks Databricks Release Notes ·
- Databricks Environment Version 6 Now Available for Serverless & Classic Compute Databricks Release Notes ·
- OpenAI GPT-6 Astra Now Available on Databricks Unity Gateway Databricks Release Notes ·