Why Kafka Keys, Not Partitions, Should Be Your Ordering Abstraction
Kafka Partitions are the wrong ordering abstraction. Keys are. Bye “Parallel Consumer” and Why the real Kafka unit of work was never the partition. Confluent has officially deprecated the …
Read the full articleYou might also wanna read
Advanced ClickHouse Partitioning Strategies for Petabyte-Scale Data Management
As datasets grow into the petabyte range, choosing the right partitioning strategy in ClickHouse becomes a critical architectural decision a
The transactional outbox pattern: keeping a database and Kafka consistent
Wrote this up after one too many "the row updated but the event never fired" incidents. It walks through why the dual-write problem is unavo
Comparing PostgreSQL and Kafka: Understanding Their Different Use Cases in Distributed Systems
Looking to make it to the front page of HackerNews? Then writing a post arguing that "Postgres is enough", or why "you don’t need Kafka at y
morling.dev·8mo agoApache Kafka® Tiered Storage in Depth: How Writes and Metadata Flow
A detailed end-to-end overview of how Kafka writes and stores metadata with KIP-405.
Apache Kafka Share Groups are NOT true queues. Here's why that's a good thing.
Kafka 4.2 Share Groups bring elastic scaling & per-message control—without strict ordering. Here's when that trade-off is exactly what you n
Does Apache Kafka® really preserve message ordering?
We’re told that Apache Kafka® preserves message ordering per topic/partition, but how true is that, and how close can we come to making it t
Comments
Sign in to join the conversation.
No comments yet. Be the first.