Connecting a visualization layer to a kafka streams / ksqlDB table?

Question

I've been fiddling around with ksqlDB and it's really neat. Getting the data into Kafka, creating streams and even tables is working out nicely. I would now like to connect a visualization layer/BI tool to the data in the tables for creating a dashboard or some interactive BI report (eg Superset, Redash, etc).

Is there a way to connect a JDBC client directly to ksqlDB / Kafka Streams so it can query the tables in ksqlDB? This video kind of says I should use the REST API for hooking up an app to the results, but:

1. The REST API is proprietary, so how would I connect something like a BI tool to it without writing a custom connecter?
2. If I run ksqlDB in production/headless mode the REST API is not available anyway. (why??)

Any pointers would be greatly appreciated.

Answer 1

Is there a way to connect a JDBC client directly to ksqlDB / Kafka Streams so it can query the tables in ksqlDB?

There is a community JDBC driver for ksqlDB.

Bear in mind that currently ksqlDB is not intended for ad-hoc random access patterns. You may want to design your streams and tables carefully to optimise for the type of queries you're running.