Re: Indexes performance - Mailing list pgsql-performance

From Josh Berkus
Subject Re: Indexes performance
Date
Msg-id 200410181245.10810.josh@agliodbs.com
Whole thread Raw
In response to Indexes performance  (charavay <c.charavay@ibcp.fr>)
List pgsql-performance
Charavay,

> ---------------------------------------------------------------------------
>-------- Hash Join  (cost=6793.29..1716853.80 rows=33743101 width=11)
>    Hash Cond: ("outer"."index" = "inner"."index")
>    ->  Seq Scan on lnk  (cost=0.00..535920.00 rows=33743100 width=15)
>    ->  Hash  (cost=4994.83..4994.83 rows=303783 width=4)
>          ->  Seq Scan dic  (cost=0.00..4994.83 rows=303783 width=4)
> (5 rows)

According to the estimate, you are selecting all of the rows in the database.
This is going to require a Seq Scan no matter what.

--
--Josh

Josh Berkus
Aglio Database Solutions
San Francisco

pgsql-performance by date:

Previous
From: Jan Wieck
Date:
Subject: Re: First set of OSDL Shared Mem scalability results, some
Next
From: Jan Wieck
Date:
Subject: Re: [HACKERS] Getting rid of AtEOXact Buffers (was Re: [Testperf-general]