Stream Processing Smackdown: Kafka Streams vs. Apache Flink

on YouTube ↗50m

" summary

Direct comparison of Kafka Streams and Flink with live-coded examples shows how each handles state, fault tolerance, and scaling: Kafka Streams uses a RocksDB-backed state store and consumer-group partitioning with optional standby replicas for faster recoveries; Flink runs jobs on a clustered JobManager/TaskManager model with incremental checkpoints and watermark-driven windows. Example pipeline computes flight delays using a 15-minute sliding window with a one-minute advance and demonstrates exactly-once semantics end-to-end when enabled.

AI-written summary · about

" chapters (12)

1. 0:00Stream processing comparison
2. 2:50Stream processing basics
3. 8:04Kafka scalability
4. 13:09Kafka Streams architecture
5. 16:40Kafka Streams demo
6. 20:53Flink architecture
7. 24:24Flink processing example
8. 29:23State management comparison
9. 32:10Scalability differences
10. 37:29Time and window ops
11. 41:06Exactly-once semantics
12. 46:07Kafka Streams vs Flink

" related talks

• 41mBeyond the Hype: When Apache Flink Solves Real ProblemsNaci Simsek · Berlin Buzzwords 2026
• 38mApache Flink: A Very Quick GuideVladimir Bacvanski · Data By the Bay 2016
• 27mNavigating Streaming InfrastructureAdi Polak · DataEngBytes Auckland 2024
• 43mA Data Streaming Architecture with Apache Flink #bbuzzRobert Metzger · Berlin Buzzwords 2016
• 49mscale.bythebay.io: Stephan Ewen, Apache Flink and the Next Wave of Stream Processing ApplicationsStephan Ewen · Scale By the Bay 2017
• 40mRobust Stream Processing with Apache FlinkJamie Grier · Flink Forward Berlin 2016
• 39mEvaluating Streaming Framework Performance for a Large-Scale...Ron Crocker · Flink Forward Berlin 2016
• 46mFlink SQL in 2020: Time to show off!Timo Walther, Fabian Hueske · Flink Forward Virtual 2020