=?windows-1250?Q?Miroslav_=8Aulc?= <miroslav.sulc@startnet.cz> writes:
> There are only JOINs number against number.
Hmph. There's no reason I can see that hash joins should be as slow as
they seem to be in your test.
Is the data confidential? If you'd be willing to send me a pg_dump
off-list, I'd like to replicate this test and try to see where the time
is going.
regards, tom lane