I see that many nodes related to data processing services are utilizing high memory. The memory usage grows from ~40% to ~85% in 20 to 60 mins and then recycles. This is happening very often. I have attached the images related to dataflow, decision hub, DDS nodes.

High memory consumption and constant recycles can be a reason for the performance problems we are facing and we also assume that it could lead to application crashes.

I would like to understand -

1) Which process on the nodes are consuming high memory?

2) Can I identify the rules associated with it? so that we can fine tune the rules.