When you are dealing with streaming data, it might seem like tables are things that dwell in the far-off land of relational databases, outside of Apache Kafka and your event streaming system. But then the Kafka Streams API gives us the KTable abstraction, which lets us create tabular views of data in Kafka topics.

Apache Kafka 2.1 featured an interesting change to the table API—commonly known to the world as KIP-328—that gives you better control over how updates to tables are emitted into destination topics. What might seem like a tiny piece of minutia gives us an opportunity to explore important parts of the Streams API, and unlocks some key new use cases. Join John Roesler for a clear explanation of the whole thing.