Unified stream and batch processing with Apache Flink’s relational APIs

on YouTube ↗43m

" summary

Apache Flink provides Table API and SQL as a single relational layer that yields identical results on files or Kafka streams by translating queries into the same logical plan via Apache Calcite and code generation. It models streams as dynamic tables, supports event-time windowing with watermarks so window results are append-only, and converts tables to streams in upsert/append modes (retractions or key-based updates). Aggregations and window joins are implemented or imminent and run in production.

AI-written summary · about

" chapters (5)

1. 0:00Introduction to stream SQL
2. 26:04Converting streams and tables
3. 30:50State and query constraints
4. 34:52Project roadmap and use cases
5. 37:46Conclusion and Q&A

" related talks

• 21mStream Processing and SQLBruno Bonnin · ScalaIO 2018
• 23mBetter, Faster, Stronger Streaming: Your First Dive into Flink SQLCaito Scherr · Berlin Buzzwords 2021
• 41mStreaming SQLJulian Hyde · Flink Forward Berlin 2016
• 42mStream Processing with Apache FlinkTugdual Grall · DevFest Toulouse 2017
• 22mStream SQL with Flink @ YelpEnrico Canzonieri · Flink Forward Europe 2019
• 51mStream Processing Smackdown: Kafka Streams vs. Apache FlinkViktor Gamov · Voxxed Days Ticino 2025
• 40mThe Seven Deadly Sins of StreamingGiannis Polyzos · Big Data Conference Europe 2025
• 31mWriting a Streaming SQL & Materialized View EngineKenny Gorman · Flink Forward Virtual 2020