Re: Horribly slow hash join

Yes, stats are up to date, and the values should be fairly unique.

Combined with the hash aggregate problem I saw (see my other email to
the list), do you think there could be some issue with the performance
of the hash function on FreeBSD 5.2 on AMD64?

I'll post the table you requested someplace you can grab it.

On Fri, Apr 16, 2004 at 12:34:11PM -0400, Tom Lane wrote:
> "Jim C. Nasby" <jim(at)nasby(dot)net> writes:
> > Note the time for the hash join step:
>
> Have you ANALYZEd these tables lately?
>
> It looks to me like it's hashing on some column that has only a small
> number of distinct values, so that the hash doesn't actually help to
> avoid comparisons. The planner should know better than to choose such
> a plan, but if it's working with obsolete stats ...
>
> regards, tom lane
>

--
Jim C. Nasby, Database Consultant jim(at)nasby(dot)net
Member: Triangle Fraternity, Sports Car Club of America
Give your computer some brain candy! www.distributed.net Team #1828

Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"