Support of streaming data sources

[navado]

Like receiving io streams and quantization criterion, Kafka and other remote queues in a similar way like Spark streaming works.

[sdpython]

It would help me if you have an example I could use to see what you would like.