Is Kafka a database?

Slides from a keynote given at Kafka Summit San Francisco, 17 October 2018.
http://martin.kleppmann.com/2018/10/17/kafka-summit.html

Abstract:

Let’s explore a contentious question: is Kafka a database? In some ways, yes: it writes everything to disk, and it replicates data across several machines to ensure durability. In other ways, no: it has no data model, no indexes, no way of querying data except by subscribing to the messages in a topic.

But there’s more to this question than meets the eye. In an organisation, data flows between different systems managed by different teams, and Kafka is becoming a popular way of realizing such data flows. In doing so, it allows us to ensure certain kinds of consistency across different systems. And if you look deeply at the structure of these data flows, it becomes possible to do new things that would have been very difficult to do otherwise.

Four years after his talk “Turning the database inside-out”, which has been watched 100,000 times, Martin delivers an update on the big ideas that connect databases and streams.

Transcript

57. Book signing Today 3.30pm Confluent booth