Thread: Query planning

Query planning

From
Jayadevan M
Date:
Hi all,

I am going through PostgreSQL documentation on how row estimation is done in PostgreSQL.
http://www.postgresql.org/docs/current/static/row-estimation-examples.html
There is a reference to cardinality (rel_cardinality) there. But that is an estimate of the number of rows in the table, right?

How many unique values are there in a column (cardinality), that is estimated while finding out selectivity?

Regards,
Jayadevan

Re: Query planning

From
Emanuel Calvo
Date:
El 31/10/14 a las 04:46, Jayadevan M escibió:
> Hi all,
>
> I am going through PostgreSQL documentation on how row estimation is
> done in PostgreSQL.
> http://www.postgresql.org/docs/current/static/row-estimation-examples.html
> There is a reference to cardinality (rel_cardinality) there. But that
> is an estimate of the number of rows in the table, right?
>
> How many unique values are there in a column (cardinality), that is
> estimated while finding out selectivity?
>
>
Did you look into pg_stats view? There is a column colled n_distinct
that indicates
that (-1 all distinct, 1 none distintct). That's what you are asking about?

--
--
Emanuel Calvo                 http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services