pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
INNER JOIN vs WHERE
2+ messages / 2 participants
[nested] [flat]

* INNER JOIN vs WHERE
@ 2003-05-15 01:45  T. Alex Beamish <talexb@tabsoft.on.ca>
  0 siblings, 1 reply; 2+ messages in thread

From: T. Alex Beamish @ 2003-05-15 01:45 UTC (permalink / raw)
  To: pgsql-performance

Hello,

I've just had a quick search through the archives but couldn't find
what I wanted, so I've joined this list to continue my search for an
answer.

Is there any rule of thumb about how much more efficient an INNER JOIN
is compare to a regular WHERE statement?

I have a couple of queries in PostgreSQL that use a variety of tables
(four or five) linked together by key/foreign key conditions all ANDed
together. My co-worker re-wrote one of them using the INNER JOIN
approach and I wanted to find out if this would empirically improve the
performance.

I have not tried to do an EXPLAIN ANALYZE yet but I will try that.
Thanks for your responses.

Alex






^ permalink  raw  reply  [nested|flat] 2+ messages in thread

* Re: INNER JOIN vs WHERE
@ 2003-05-15 02:25  Tom Lane <tgl@sss.pgh.pa.us>
  parent: T. Alex Beamish <talexb@tabsoft.on.ca>
  0 siblings, 0 replies; 2+ messages in thread

From: Tom Lane @ 2003-05-15 02:25 UTC (permalink / raw)
  To: T. Alex Beamish <talexb@tabsoft.on.ca>; +Cc: pgsql-performance

"T. Alex Beamish" <talexb@tabsoft.on.ca> writes:
> Is there any rule of thumb about how much more efficient an INNER JOIN
> is compare to a regular WHERE statement?

Ideally there is no difference.  If there are only two tables involved,
there definitely is no difference.

If there is a difference, it's because there are more than two tables,
and the JOIN syntax forced a particular join order.  Read 
http://www.postgresql.org/docs/view.php?version=7.3&idoc=0&file=explicit-joins.html

Typically I'd expect JOIN syntax to be a loss because the odds are good
that you're forcing an inefficient join order.  It could be a win only
if the planner chooses a poor join order when given a free hand, or if
you have so many tables that you need to suppress the planner's search
for a good join order.

> I have not tried to do an EXPLAIN ANALYZE yet but I will try that.

If you have not bothered to do any EXPLAINs yet then you are really
wasting people's time on this list.

			regards, tom lane



^ permalink  raw  reply  [nested|flat] 2+ messages in thread


end of thread, other threads:[~2003-05-15 02:25 UTC | newest]

Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2003-05-15 01:45 INNER JOIN vs WHERE T. Alex Beamish <talexb@tabsoft.on.ca>
2003-05-15 02:25 ` Tom Lane <tgl@sss.pgh.pa.us>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox