Dear Tomas, > 1) I really like the idea of introducing "compute_workers" callback to > the heap AM interface. I faced a similar issue with calculating workers > for index builds, because right now plan_create_index_workers is doing > that the logic works for btree, but really not brin etc. It didn't occur > to me we might make this part of the index AM ...
+1, so let's keep the proposed style. Or, can we even propose the idea to table/index access method API? I've considered bit and the point seemed that which arguments should be required. > 4) I think it would be good to have some sort of README explaining how > the parallel heap vacuum works, i.e. how it's driven by FSM. Took me a > while to realize how the workers coordinate which blocks to scan. I love the idea, it is quite helpful for reviewers like me. Best regards, Hayato Kuroda FUJITSU LIMITED
