Strimzi is a Kubernetes operator for running Apache Kafka. It turns every moving part of a Kafka deployment, brokers, controllers, topics, users, connectors, and mirror pipelines, into Kubernetes custom resources, so operators can kubectl apply a cluster rather than shepherd JVMs and ZooKeeper ensembles by hand.
The project ships several CRDs and controllers. Kafka describes a cluster and its listeners, storage, authentication, and authorization; the Cluster Operator then creates StatefulSets, Services, and ConfigMaps to match. KafkaTopic and KafkaUser are reconciled by dedicated Entity Operators that drive changes into Kafka via its admin API. KafkaConnect, KafkaConnector, KafkaMirrorMaker2, and KafkaBridge cover connect pipelines, cross-cluster replication, and HTTP access. Strimzi supports both classic ZooKeeper mode and KRaft mode and handles rolling upgrades, rebalancing via Cruise Control, TLS cert rotation, and OAuth integration.
Strimzi is a CNCF Incubating project originated by Red Hat and is the upstream for Red Hat’s AMQ Streams and OpenShift Streams for Apache Kafka. It is the dominant way to run self-managed Kafka on Kubernetes.