Re: Improving the Performance of Full Table Updates

"Gokulakannan Somsundaram" <gokul007@gmail.com> writes:
> I propose to change this row-by-row approach, when it is a full table
> update. I plan to send a extra flag(which will be set for Full table
> Deletes/Updates). this would make the access method directly acquire the
> exclusive lock and update the existing record.

This sounds like a recipe for utter destruction of what little
modularity and layering we've got left.  And I rather doubt it will buy
anything interesting performance-wise.

To cite just one concrete objection: surely the tuple-fetch code has got
no business firing triggers.
        regards, tom lane