Change data capture ingestion

Preview

This feature is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms. Pre-GA features are available "as is" and might have limited support. For more information, see the launch stage descriptions.

Change data capture (CDC) ingestion lets you synchronize your source databases with borderless Lakehouse Apache Iceberg managed tables (formerly BigLake tables for Apache Iceberg in BigQuery) in real time. By capturing incremental changes—such as inserts, updates, and deletes—CDC ensures that your lakehouse reflects the most current state of your operational data without requiring full table reloads.

Apache Iceberg managed tables use BigQuery CDC concepts and capabilities to efficiently process these changes and enable low-latency analytics on data stored in open formats like Apache Iceberg. This approach significantly reduces the management overhead associated with typical ETL processes.

Limitations

Apache Iceberg managed tables support change data capture (CDC) ingestion, and leverage the same concepts and capabilities as standard BigQuery CDC ingestion. However, there are a few additional limitations to be aware of:

• Out-of-order deletions are not supported. Unlike BigQuery CDC ingestion, a delete retention window is not maintained for CDC-enabled Iceberg managed tables. If a DELETE event arrives before the corresponding record is inserted, the deletion is discarded by the background apply jobs. This can result in out of order records remaining in your table that were intended to be deleted.
• All limitations under Apache Iceberg managed tables limitations.
• All limitations under BigQuery CDC ingestion limitations.

What's next

For more information, see the following BigQuery documentation:

• Stream data with change data capture
• Stream data with the BigQuery Storage Write API (gRPC)
• Learn how to use Datastream to replicate transactional databases to BigLake tables.