Re: IN vs EXISTS equivalence - Mailing list pgsql-hackers

From Kevin Grittner
Subject Re: IN vs EXISTS equivalence
Date
Msg-id 48983AA7.EE98.0025.0@wicourts.gov
Whole thread Raw
In response to Re: IN vs EXISTS equivalence  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: IN vs EXISTS equivalence  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
>>> On Mon, Aug 4, 2008 at  6:48 PM, in message
<15026.1217893692@sss.pgh.pa.us>,
Tom Lane <tgl@sss.pgh.pa.us> wrote:
> "Kevin Grittner" <Kevin.Grittner@wicourts.gov> writes:
>> I'm adding some NOT EXISTS examples to the thread for completeness
of
>> what someone might want to address while working on it.  For two
>> queries which can easily be shown (to a human viewer, anyway) to
>> return identical results, I see performance differences of over
five
>> orders of magnitude.
>
> Could we see EXPLAIN ANALYZE not just EXPLAIN for these?  When
people
> are complaining of bad planner behavior, I don't find bare EXPLAIN
> output to be very convincing.

The other five queries have a cost to millisecond ratio of between 9.8
and 267.  If the expensive one falls in the same range, it will run
for 2.3 to 64 years.  I know I have left it running for days before
without completion.  I don't think I can devote the resources to it.
Attached are the EXPLAIN ANALYZE output for the other five.

-Kevin

Attachment

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Status of DISTINCT-by-hashing work
Next
From: Stephen Frost
Date:
Subject: Re: Parsing of pg_hba.conf and authentication inconsistencies