Hi thanks for the reply ,i guess there was some ongoing work for this and for some reason it didn't continued this is the thread https://www.postgresql.org/message-id/flat/CAJcOf-cXnB5cnMKq... if you are interested in, but for sure i will look into this(added to my list:).

What can I say, except thank you very much.

That seems like an unintuitive consequence. Saturating disk would have been obvious. But CPU?

Does anything about the INSERT pattern seem "special"? Or does it happen for many different "kinds of INSERT statements"?