Thread: Incorrect estimates on correlated filters

On Aug 12, 2008, at 4:59 PM, Chris Kratz wrote:
> Ran into a re-occuring performance problem with some report queries
> again today.  In a nutshell, we have filters on either multiple
> joined tables, or multiple columns on a single table that are
> highly correlated.  So, the estimates come out grossly incorrect
> (the planner has no way to know they are correlated).  2000:1 for
> one I'm looking at right now.  Generally this doesn't matter,
> except in complex reporting queries like these when this is the
> first join of 40 other joins.  Because the estimate is wrong at the
> lowest level, it snowballs up through the rest of the joins causing
> the query to run very, very slowly.   In many of these cases,
> forcing nested loops off for the duration of the query fixes the
> problem.  But I have a couple that still are painfully slow and
> shouldn't be.
>
> I've been reading through the archives with others having similar
> problems (including myself a year ago).  Am I right in assuming
> that at this point there is still little we can do in postgres to
> speed up this kind of query?  Right now the planner has no way to
> know the correlation between different columns in the same table,
> let alone columns in different tables.  So, it just assumes no
> correlation and returns incorrectly low estimates in cases like these.
>
> The only solution I've come up with so far is to materialize
> portions of the larger query into subqueries with these correlated
> filters which are indexed and analyzed before joining into the
> larger query.  This would keep the incorrect estimates from
> snowballing up through the chain of joins.
>
> Are there any other solutions to this problem?


Well... you could try and convince certain members of the community
that we actually do need some kind of a query hint mechanism... ;)

I did make a suggestion a few months ago that involved sorting a
table on different columns and recording the correlation of other
columns. The scheme isn't perfect, but it would help detect cases
like a field populated by a sequence and another field that's insert
timestamp; those two fields would correlate highly, and you should
even be able to correlate the two histograms; that would allow you to
infer that most of the insert times for _id's between 100 and 200
will be between 2008-01-01 00:10 and 2008-01-01 00:20, for example.
--
Decibel!, aka Jim C. Nasby, Database Architect  decibel@decibel.org
Give your computer some brain candy! www.distributed.net Team #1828

Chris Kratz wrote:

> Unfortunately, if I don't think the sorting idea would help in the one case
> I'm looking at which involves filters on two tables that are joined
> together.  The filters happen to be correlated such that about 95% of the
> rows from each filtered table are actually returned after the join.
> Unfortunately, the planner thinks we will get 1 row back.

Maybe you can wrap that part of the query in a SQL function and set its
estimated cost to the real values with ALTER FUNCTION ... ROWS.

--
Alvaro Herrera                                http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support

Decibel! wrote:

> Well... you could try and convince certain members of the community that
> we actually do need some kind of a query hint mechanism... ;)

It strikes me that there are really two types of query hint possible here.

One tells the planner (eg) "prefer a merge join here".

The other gives the planner more information that it might not otherwise
have to work with, so it can improve its decisions. "The values used in
this join condition are highly correlated".

Is there anything wrong with the second approach? It shouldn't tend to
suppress planner bug reports etc. Well, not unless people use it to lie
to the planner, and I expect results from that would be iffy at best. It
just provides information to supplement Pg's existing stats system to
handle cases where it's not able to reasonably collect the required
information.

--
Craig Ringer

"Craig Ringer" <craig@postnewspapers.com.au> writes:

> It strikes me that there are really two types of query hint possible here.
>
> One tells the planner (eg) "prefer a merge join here".
>
> The other gives the planner more information that it might not otherwise
> have to work with, so it can improve its decisions. "The values used in
> this join condition are highly correlated".

This sounds familiar:

http://article.gmane.org/gmane.comp.db.postgresql.devel.general/55730/match=hints

Plus ça change...

--
  Gregory Stark
  EnterpriseDB          http://www.enterprisedb.com
  Ask me about EnterpriseDB's On-Demand Production Tuning

On Aug 13, 2008, at 1:45 PM, Chris Kratz wrote:
> Yes, I know hints are frowned upon around here.  Though, I'd love
> to have them or something equivalent on this particular query just
> so the customer can run their important reports.  As it is, it's
> unrunnable.


Actually, now that I think about it the last time this was brought up
there was discussion about something that doesn't force a particular
execution method, but instead provides improved information to the
planner. It might be worth pursuing that, as I think there was less
opposition to it.
--
Decibel!, aka Jim C. Nasby, Database Architect  decibel@decibel.org
Give your computer some brain candy! www.distributed.net Team #1828