News, guides, and engineering deep dives.
Practical guidance on Kafka, Flink, Iceberg, and real-time data.
Kafka deployment automation
Deployment automation for Kafka: zero-downtime rolling changes, declarative topics and ACLs, Kubernetes operators, Terraform and GitOps for cluster state.
The difference between Kafka and RabbitMQ
The difference between Kafka and RabbitMQ is the data model: a replayable log against a delete-on-acknowledge queue. Storage, routing, scaling and when to pick which.
How RabbitMQ works
How RabbitMQ works, mapped to Kafka concepts: exchanges and bindings, per-message acknowledgement, competing consumers, quorum queues and dead lettering.
Kafka ACL
A Kafka ACL allows or denies a principal an operation on a resource from a host. Syntax, copy-paste commands, production best practices and GitOps automation.
Kafka authentication
Kafka authentication verifies every client and broker connection with SASL or mutual TLS. Listener and JAAS blueprints, mechanism choice, rotation and troubleshooting.
Kafka consumer groups
A Kafka consumer group shares the work of consuming a topic, one partition per member. Troubleshooting lag and rebalances, the CLI cheat sheet, coordinator architecture and offset commits.
Kafka Connect
Kafka Connect moves data between Kafka and external systems through source and sink connectors. What a deployment is made of, the operational tasks, and where the sub-pages go deeper.
Kafka Connect MongoDB example
A production Kafka Connect MongoDB example in both directions: source connector via change streams, sink connector with idempotent writes, DLQ settings and secrets handled properly.
Kafka Connect pricing
Kafka Connect is free open source software. The cost is in running it: managed per-task and per-GB billing, network surcharges, or the engineering hours of self-hosting. The TCO math, honestly.