From: Amit Kapila <amit.kapil...@gmail.com> > It is quite possible what you are saying is correct but I feel that is > not this patch's fault. So, won't it better to discuss this in a > separate thread? > > Good use case but again, I think this can be done as a separate patch.
Agreed. I think even the current patch offers great benefits and can be committed in PG 14, even if all my four feedback comments are left unaddressed. I just touched on them for completeness in terms of typically expected use cases. They will probably be able to be implemented along the current design. > I think here you are talking about the third patch (Parallel Inserts). > I guess if one has run inserts parallelly from psql then also similar > behavior would have been observed. For tables, it might lead to better > results once we have the patch discussed at [1]. Actually, this needs > more investigation. > > [1] - > https://www.postgresql.org/message-id/20200508072545.GA9701%40telsaso > ft.com That looks interesting and worth a try. Regards Takayuki Tsunakawa
