Greg Stark <stark@mit.edu> writes:
> I have a sketch for how to handle spilling hash aggregates to disk in
> my head. I'm not sure if it's worth the amount of complexity it would
> require but I'll poke around a bit and see if it works out well.
It'd be awfully nice if those could spill to disk. I think that
currently that's the only plan type where a misestimate can lead to
hard failure rather than just slower-than-you'd-like. Which is not
nice considering that the estimates are necessarily just estimates.
Could you give us a brain dump on the sketch? I've never seen how to
do it without unreasonable overhead.
regards, tom lane