|From:||Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>|
|To:||PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>|
|Subject:||Using multiple extended statistics for estimates|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
PostgreSQL 10 introduced extended statistics, allowing us to consider
correlation between columns to improve estimates, and PostgreSQL 12
added support for MCV statistics. But we still had the limitation that
we only allowed using a single extended statistics per relation, i.e.
given a table with two extended stats
CREATE TABLE t (a int, b int, c int, d int);
CREATE STATISTICS s1 (mcv) ON a, b FROM t;
CREATE STATISTICS s2 (mcv) ON c, d FROM t;
and a query
SELECT * FROM t WHERE a = 1 AND b = 1 AND c = 1 AND d = 1;
we only ever used one of the statistics (and we considered them in a not
particularly well determined order).
This patch addresses this by using as many extended stats as possible,
by adding a loop to statext_mcv_clauselist_selectivity(). In each step
we pick the "best" applicable statistics (in the sense of covering the
most attributes) and factor it into the oveall estimate.
All this happens where we'd originally consider applying a single MCV
list, i.e. before even considering the functional dependencies, so
roughly like this:
... apply another MCV list ...
... apply functional dependencies ...
I've both in the loop, but I think that'd be wrong - the MCV list is
expected to contain more information about individual values (compared
to functional deps, which are column-level).
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
|Next Message||Robert Haas||2019-10-28 15:28:49||Re: WIP/PoC for parallel backup|
|Previous Message||Tom Lane||2019-10-28 15:20:44||Re: Proposition to use '==' as synonym for 'IS NOT DISTINCT FROM'|