I have few question regarding apache beam. if apache beam is using its own stateful processing then which runner will be used?
if we are using Java SDK which runner is flexible whether is classic apache flink runner or portable flink runner. Suppose we are using apache flink runner who does take care take care of stateful processing for fault tolerance. How is Back pressure is handled in Apace beam. if we integrate apache beam with flink and then use apache flink runner what are the overheads? Could you please response at the earliest kindly would be helpful. Thanks Azhar
