Alvaro Herrera <alvherre@commandprompt.com> writes:
> Huh, but the log would not be flushed for each operation that the vacuum
> logs. Only when it's going to commit.
It strikes me that the vacuum cost delay feature omits to consider
generation of WAL records as a cost factor. It may not be a big problem
though, as long as we can limit the number of records created to one or
two per page --- then you can see it as just a component of the "dirtied
a page" cost. If we made a separate WAL record for each tuple then it
could be important to account for.
regards, tom lane