Event streaming, pub/sub, backpressure, and consumer groups at scale.
An article joins this feed when it matches these terms. Each one is also a search of its own.
There's a new JDBC driver that provides standard access to Apache Kafka via JDBC API. The current version connects to the KSQL engine to perform queries to Kafka and then, the engine translates those requests to Kafka requests. The sql-jdbc-driver…
Six consumers cleared a 300,000-message backlog in 5.1 s, and eight were never faster. A worker pool reordered … Kafka Consumer Lag Keeps Growing: Backlog, Ordering and Priority, 4 Scenarios Measured Six consumers cleared a 300,000-message backlog in 5.1 s, and eight were never…
Kafka is mandatory for any Backend / Java / Microservices interview. This is Part 1 of a two-part prep sheet — it covers the core architecture, partitioning, brokers, and delivery semantics questions that come up in almost every interview. Part 2 moves into real-world design…
On February 17, 2026, Apache Kafka 4.2 shipped with this line in the release announcement: "Kafka Queues (Share Groups) is now production-ready." That sentence quietly reopens one of backend engineering's oldest arguments. For years, "should my job queue be Postgres or Kafka?" had…
ZooKeeper is rarely the headline system in an enterprise stack. It sits underneath one. Apache Hadoop and Apache Kafka both rely on ZooKeeper to manage cluster state. When the coordination layer is weak, the applications above it inherit that weakness.
Your application writes to a database. Meanwhile, other people want that data somewhere else: analytics wants it in a warehouse, search wants it in Elasticsearch, a cache needs invalidating, and a downstream microservice needs to react when an order ships.
Testing an application on Ubuntu can involve sending a message from one program and reading it in another. Apache Kafka stores messages from applications for other applications to read later. I’ll show you how to download and install Kafka on…
In this deep architectural post-mortem, we analyze the root cause of PID churn and provide production-hardened configurations to maintain deterministic EOS. When an idempotent Kafka producer initializes without a static transactional ID, the broker's Transaction Coordinator assigns a new 64-bit…
Live Webinar and Q&A: When AI Accelerates Development, Can Your CI Pipeline Keep Up? (Oct 8, 2026) Save Your Seat Susan Chang explains how Elastic transitioned from siloed, ad-hoc AI agent evaluations to a unified, production-grade framework. She discusses balancing LLM-as-a-judge with…
OpenLogic experts will assess your infrastructure and advise you on next steps for migrating from proprietary software to open source alternatives. The Open Source Digital Autonomy Playbook provides a practical framework for assessing cloud dependencies and reducing vendor lock-in to…