Kafka cluster
A Kafka cluster is a group of Kafka brokers that collectively store, distribute, and serve event streams.
Apache Kafka in Action
A Kafka cluster is a group of Kafka broker nodes that jointly store and serve event data, with security controls determining how clients and brokers connect and access it.
- How it is used
- Serves as the distributed event backbone connecting producers and consumers through topics and brokers.
- How deep it goes
- Works through replication, coordination, metadata-based connections, and broker-count trade-offs, while leaving server provisioning unspecified.
- Examples
- Uses bank-transfer events, a three-broker topic-listing command, and separate clusters for performance, security, or geography.
- What you can do after
- Operate/configure it with multiple bootstrap addresses, plan at least three brokers for resilience, and assess whether separate clusters are warranted.
Kafka for Architects
A Kafka cluster is a group of broker nodes operating together to provide distributed event storage, processing, and service availability.
- How it is used
- Explains how Kafka distributes messaging, storage, coordination, and availability across cooperating broker nodes.
- How deep it goes
- Descends to replication, acknowledgments, consumer recovery, metadata coordination, and operational trade-offs, but stops short of a complete operations guide.
- Examples
- Uses producer acknowledgments, retries, broker failure, consumer reconnection, message replay, and broker-controller architecture diagrams.
- What you can do after
- Operate/configure a cluster by evaluating capacity, monitoring, scaling, upgrades, testing, and disaster-recovery requirements.
Related concepts
- Kafka brokerhas part · A Kafka broker is a server process in an Apache Kafka cluster that stores records and handles producer, consumer, and other client requests.
- message brokerhas part · A message broker is intermediary software that receives, routes, and delivers messages between communicating producers and consumers.
- controllerhas part · A Kafka controller is a server role that manages cluster metadata and coordinates partition, replica, and leader state.
- bootstrap serverconfigured by · A bootstrap server is an initial broker endpoint that a Kafka client uses to connect to a cluster and retrieve metadata.
- GitOpsconfigured by · GitOps is an operational approach that uses version-controlled declarative configuration as the source of truth and automation to apply or reconcile system state.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.