Connect

Change Data Capture (CDC)

Redpanda Connect captures row-level changes from databases in real time and streams them to Redpanda topics. CDC inputs connect directly to database transaction logs or change streams, transforming each change into a message that downstream applications can process immediately.

After reading this page, you will be able to:

  • Describe how CDC inputs capture database changes in real time

  • Compare Redpanda Connect CDC with Kafka Connect approaches

  • Identify which CDC input to use for your database

Common use cases

CDC enables real-time data integration patterns:

  • Sync data between databases and data warehouses continuously, eliminating batch ETL delays

  • Build event-driven microservices that react to data changes within milliseconds

  • Replicate data across regions automatically for disaster recovery

  • Feed real-time analytics pipelines from operational databases without impacting performance

  • Keep search indexes synchronized with primary data stores without polling

CDC inputs in Redpanda Connect

Redpanda Connect CDC inputs use declarative YAML configuration and integrate with Bloblang transformations for data manipulation. This approach eliminates the need for a separate Kafka Connect cluster, and handles schemas through built-in schema registry processors.

Feature Redpanda Connect CDC Kafka Connect

Configuration

Declarative YAML

JSON + REST API

Transformations

Native Bloblang

SMTs (Single Message Transforms)

Schema evolution

Built-in Schema Registry support

Requires additional configuration

Deployment

Single binary, managed in Redpanda Cloud

Separate cluster required

For managed Redpanda Cloud deployments, CDC pipelines run as fully managed services. For self-hosted deployments, configure CDC inputs using Redpanda Connect configuration files.

Supported databases

Redpanda Connect provides CDC inputs for relational databases, NoSQL databases, and cloud services.

Relational databases

NoSQL databases

  • MongoDB CDC - Change streams for replica sets and sharded clusters

  • DynamoDB CDC - DynamoDB Streams for real-time table changes

Cloud services

Example patterns and configurations

The following cookbooks demonstrate real-world CDC patterns with complete pipeline configurations.

Oracle CDC pipeline

See Oracle CDC patterns for complete examples of filtering, transforming, and routing Oracle database changes.

DynamoDB CDC pipeline

See DynamoDB CDC patterns for patterns that stream DynamoDB table changes to Redpanda.

Suggested reading