Optimizing Kafka Streams Applications | Confluent

This post introduces the processor topology optimization framework in Kafka Streams (since version 2.1.0). It explains how processor topologies are generated from the Processor API and Streams DSL, illustrates the use of Topology#describe() and external visualization tools, and details the concept of data parallelism and repartition topics for stateful operations. It highlights how the framework allows for optimizations by analyzing the topology graph and introduces techniques like sticky partitioning for improved efficiency.