The current execution model for KNIME is that every node is executed in a job. Jobs can be executed in parallel but if the flow is split.
But is there a way to execute the flow and have all nodes executed simultaneously. That is can the nodes all be started and have them process data as soon as the upstream node has produced a single row. Not have to way for all the rows to be produced before kicking off the subsequent node.
From looking at the code it seems that this could be achieved if a synchronized TableData existed and if nodes returned their TableData and then wrote into them while subsequent nodes read from them.
Is this something being considered or planed for?