Kafka Streams Flashcards

What type of library is Kafka Streams?

Answer: A client-side Java library for stream processing, requiring no separate cluster

Kafka Streams is a lightweight Java library embedded in your application — it uses Kafka topics as input/output and needs no additional infrastructure.

What is a KStream in Kafka Streams?

Answer: An unbounded sequence of individual key-value records (changelog stream)

A KStream represents a continuous, unbounded stream of key-value records where each record is an independent event.

What is a KTable in Kafka Streams?

Answer: A changelog stream interpreted as a table where each record is an upsert by key

A KTable models a mutable table backed by a changelog topic; each new record with an existing key updates (upserts) the current value for that key.

Which method is used to materialize a KTable into a queryable local state store?

Answer: Materialized.as()

Materialized.as() specifies a named state store backed by RocksDB where the KTable's data is accessible for interactive queries.

What is the default state store backend used by Kafka Streams for persistent state?

Answer: RocksDB

Kafka Streams uses RocksDB as its default embedded persistent key-value store for local state in aggregations and joins.

What does a 'stream-table join' in Kafka Streams do?

Answer: Enriches each stream record with the current table value for the matching key

A stream-table join looks up the latest KTable value for a record's key and enriches the stream record with that value in real time.