Re: Unexpected sequence scan

From: Kevin Grittner
Subject: Re: Unexpected sequence scan
Date: ,
Msg-id: 4FA3A410020000250004787C@gw.wicourts.gov
(view: Whole thread, Raw)
In response to: Unexpected sequence scan  (Dan Fairs)
List: pgsql-performance

Tree view

Unexpected sequence scan  (Dan Fairs, )
 Re: Unexpected sequence scan  ("Kevin Grittner", )
 Re: Unexpected sequence scan  (Tom Lane, )
  Re: Unexpected sequence scan  (Dan Fairs, )

Dan Fairs <> wrote:

> I have a query which is running slowly, and the query plan shows
> an unexpected sequence scan where I'd have expected the planner to
> use an index.

Looking at the actual row counts compared to run time, it appears
that the active portion of your data set is heavily cached.  In such
an environment, I would add these lines to postgresql.conf, to
better model costs:

seq_page_cost = 0.1
random_page_cost = 0.1  # or maybe slightly higher
cpu_tuple_cost = 0.03

-Kevin


pgsql-performance by date:

From: Richard Jones
Date:
Subject: Re: Partitioned/inherited tables with check constraints causing slower query plans
From: Martin Grotzke
Date:
Subject: Re: Several optimization options (config/hardware)