Change Data Streaming Patterns in Distributed Systems
summary
Shows concrete patterns and working code: CDC with Debezium + Kafka Connect reads Postgres transaction logs, routes outbox rows into per-aggregate Kafka topics, and uses SMTs and Kafka Streams to reshape and join records. Migration uses a proxy plus CDC to incrementally extract a customer service (strangler fig). Saga orchestration uses an outbox table per service and message-driven compensations so failed payments trigger compensating actions; demo runs in Docker Compose and the example completes and compensates as expected.
AI-written summary · about
chapters (7)
- 0:00Introduction to CDC
- 1:36Understanding CDC mechanics
- 4:13The outbox pattern
- 7:44Strangler fig pattern
- 14:46Advanced CDC pipelines
- 17:56Saga pattern and demo
- 27:24Conclusion and wrap-up
related talks
41mChange Data Streaming Patterns in Distributed Systems
36mChange Data Streaming Patterns for Distributed Systems
41mStreaming your database - Easier said than done?
52mChange Data Streaming For Spring Boot With Apache Kafka and Debezium
37mData Contracts In Practice With Debezium and Apache Flink
23mChange Data Capture: A RabbitMQ and Debezium Approach
42mData Streaming for Microservices using Debezium
18mData Modernization Using Confluent and Yugabyte