Apache Kafka was originally developed at LinkedIn by Jay Kreps, Neha Narkhede, and Jun Rao in 2010. LinkedIn needed a unified system for handling real-time data feeds from their website. The name “Kafka” honors the author Franz Kafka, known for his writings about bureaucracy and complex systems - fitting for a distributed messaging system.
| Year | Milestone |
|---|---|
| 2012 | Entered Apache Incubator |
| 2012 | Became Apache Top-Level Project |
| 2014 | Kafka 0.8 with replication support |
| 2015 | Kafka 0.9 with security features |
In 2014, the original Kafka creators founded Confluent to provide commercial support and enterprise features:
The most significant architectural change in Kafka’s history:
| Version | Change |
|---|---|
| 2.8.0 (2021) | KRaft mode introduced (preview) |
| 3.0.0 (2022) | KRaft production-ready |
| 3.3.0 (2023) | KRaft becomes default |
| 4.0.0 (2024) | ZooKeeper removed entirely |
KRaft (Kafka Raft) replaces ZooKeeper with an internal consensus protocol, simplifying deployment and improving scalability.
| Aspect | Status |
|---|---|
| Latest Version | 4.2.0 (2026) |
| License | Apache-2.0 |
| Architecture | KRaft only (ZooKeeper removed) |
| Use Cases | Event streaming, data pipelines, log aggregation |
| Year | Event |
|---|---|
| 2010 | Development begins at LinkedIn |
| 2011 | Open source release |
| 2012 | Apache Top-Level Project |
| 2014 | Confluent founded |
| 2021 | KRaft mode introduced |
| 2024 | Kafka 4.0 removes ZooKeeper |
| 2026 | Kafka 4.2 stable |