the program is composed by:

6 Kafka /source/ connector with custom timestamp and watermark /extractor/
and /map/ function each.
then I use 6 instance of an external library called flink-htm (quite heavy)
moreover I have 6 /process/ method and 2 /union/ method to merge result
Finally I have 2 Cassandra /sinks/.

Data which arriving to kafka are 1 kb strings about each 20ms.

I'm absolutely sure that the flink-htm library is heavy but I hoped flink
managed them distributing the load (which are independent) through the
cluster... instead it seems like just one node suffers all the load

If can help I can share my code.

Sent from: