Automatic change data feed (Auto CDF)
GALast seen at this status: 2026-09-01·45d ago
Computes row-level changes at query time from row tracking, so change data feed no longer has to be enabled on each Delta Lake or Apache Iceberg v3 table. Removing the write-time overhead makes MERGE and UPDATE about 15% faster on tables queried for changes, and it requires Databricks Runtime 19 or above with row tracking enabled. It is rolling out to workspaces and is due in all supported regions by the end of October 2026.
Status history
- GA
Sources (1)
- 2026-09-01Release notes GA
Recent coverage
- 2026-10-12v1.12.7 release
- 2026-10-06Scaling and Operating a Large dbt Project on Databricks: IFCO's Data Team on Performance, Visibility, and Debugging news
- 2026-10-05Delta Lake 4.4.1 release
- 2026-10-05NEAREST BY Join: Scaling Vector Search in Databricks Runtime news
- 2026-10-05Unlocking Data Portability: Preventing Catalog Lock-in with REGISTER and UNREGISTER APIs news
- 2026-09-28rust-v1.0.0 release
Status and dates are extracted from public sources only. We don't see the private AE-shared roadmap. If something looks wrong, tell us.