Flink SQL: The Challenges to Build a Streaming SQL Engine

Explore the challenges of building a modern streaming SQL engine like Flink SQL in this 45-minute conference talk from Confluent. Dive into the complexities of handling late data, ingesting and processing change data in real-time, maintaining event ordering during shuffling, addressing nondeterminism in functions and external system lookups, and effectively managing state storage for infinite datasets. Learn how Flink SQL resolves these issues to guarantee result correctness and see real-world examples of solving common stream processing problems. Gain valuable insights into the intricacies of streaming SQL engines and their applications in data-driven organizations.