Data generation for stream targets

[vgkowski]

The BatchReplayer only generates data in batch or micro-batch and writes to batch type targets. We need to allow for writing to streams including Amazon Kinesis Streams and Amazon MSK

If the BatchReplayer cannot be used, we should implement a new Construct, long running.

[nicusX]

It would be great if generated streams were also joinable, over time windows (i.e. multiple elements on both side can join, but only few join over limited time windows)

[vgkowski]

@nicusX your need is more related to the dataset that is replayed than the data generator in itself. The generator will just take an existing dataset and replay it. I think we are already doing this with the retail dataset in the AWS native reference architecture
If you have a specific request for a particular dataset please open another issue