I'm researching the possibility of using Kafka as the main storage for an event sourcing pattern. I'm having a hard time understanding if it is a good idea to store things in Kafka, longer term, or why not.
What would be the implications of simply setting log.retention.hours to a very large number, effectively turning Kafka into a permanent storage? As I've understood it - "Kafka's performance is effectively constant with respect to data size so retaining lots of data is not a problem."
That said, I also get the sense that this is not a common use case for Kafka, so there might be some limitation that I'm not understanding. I'm completely open to this being a bad idea, but I would like to understand why.
