Re: Planner constants for RAM resident databases

From: Emil Briggs <emil(at)baymountain(dot)com>
To: John A Meinel <john(at)arbash-meinel(dot)com>
Cc: Postgresql Performance <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Planner constants for RAM resident databases
Date: 2005-07-02 13:44:07
Message-ID: 200507020944.07339.emil@baymountain.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

> When you do "explain analyze" of a query that you have difficulties
> with, how are the planner's estimates. Are the estimated number of rows
> about equal to the actual number of rows?

Some of them are pretty far off. For example

-> Merge Left Join (cost=9707.71..13993.52 rows=1276 width=161) (actual
time=164.423..361.477 rows=49 loops=1)

I tried setting enable_merge_joins to off and that made the query about three
times faster. It's using a hash join instead.

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Tom Lane 2005-07-02 13:54:40 Re: B-Tree index not being used
Previous Message Dennis Bjorklund 2005-07-02 05:24:26 Re: planner picking more expensive plan