Debezium
By Red Hat
Debezium is an open-source distributed platform for change data capture (CDC) that captures row-level changes in databases and streams them as event logs, typically via Apache Kafka.
Definition
Debezium is an open-source distributed platform for change data capture (CDC) that captures row-level changes in databases and streams them as event logs, typically via Apache Kafka.
Overview
Debezium watches a database's transaction or replication log directly, rather than repeatedly polling tables for changes, which lets it capture inserts, updates, and deletes as an ordered, low-latency stream of events as they happen. It's built as a set of source connectors for Apache Kafka Connect, with connectors available for databases including PostgreSQL, MySQL, MongoDB, SQL Server, and Oracle. Each connector reads the database's native replication mechanism — for example, the binlog in MySQL or the write-ahead log (WAL) in PostgreSQL — and can also perform an initial snapshot of existing data before switching to streaming ongoing changes. Debezium is a common building block for event-driven architectures: streaming changes into Kafka to keep search indexes like Elasticsearch in sync, feeding real-time analytics pipelines, or replicating data between systems without batch ETL delays. Understanding this kind of streaming pipeline is a natural extension of the concepts in SkillVeris's Apache Kafka & Messaging course.
Key Features
- Log-based change data capture — reads transaction/replication logs rather than polling
- Kafka Connect source connectors for MySQL, PostgreSQL, MongoDB, SQL Server, Oracle, and more
- Captures inserts, updates, and deletes as ordered, structured events
- Initial snapshotting of existing data before streaming ongoing changes
- Schema change tracking alongside data changes
- Configurable delivery semantics depending on connector and setup
- Open source and vendor-neutral, deployable via Kafka Connect or as an embedded engine