Cassandra + kafka for event sourcing

Question

I've been reading for a while about event sourcing and CQRs and was trying to find posts/readings about using cassandra as my event store and kafka as the queue to publish events. most event sourcing posts refer to eventStore and don't give a clear discussion about cassandra+kafka combination.

Does anyone know good reading material regarding that stack? Or even answer advantages/disadvantages of using that combination? It seems cassandra is considered for read part of cqrs but no examples/data models about using it for persisting event sourced aggregates

Answer 1

As far as I can tell, Kafka wasn't designed with Event Sourcing or CQRS in mind in the first place so maybe that's why you don't find much on the subject.

Now Martin Kleppmann gave a talk about Kafka + Event Sourcing at DDD Europe 2016. His slides are mostly schemas and drawings, but maybe you can make sense of them. I know that the session was recorded though so you might want to check http://dddeurope.com/2016/ in the following months for the video.

Answer 2

Old question but you should definitely check lagomframework.com/documentation/

Its exactly the stack you are talking about.

Answer 3

Hi hsen I'm also interested in that subject and for what I have seen so far you would need a piece of software that could serve both as event-store and an event-bus at the same time for the sake of the atomic operation so you would avoid any inconsistency in any state change operation.

In my findings in the Open Sourced frameworks seems that the Akka framework could play that role using Cassandra as a persistence store.

Maybe the Apache Kafka could also be other option that could be integrated with Akka as well.