Skip site navigation (1) Skip section navigation (2)

Re: PostgreSQL Tweaking

From: Alastair Turner <bell(at)ctrlf5(dot)co(dot)za>
To: renier <renier(at)vvconsult(dot)co(dot)za>
Cc: jnbpug(at)postgresql(dot)org
Subject: Re: PostgreSQL Tweaking
Date: 2010-04-09 11:13:26
Message-ID: (view raw, whole thread or download thread mbox)
Lists: jnbpug
On Fri, Apr 9, 2010 at 12:54 PM, renier <renier(at)vvconsult(dot)co(dot)za> wrote:
> Explain Analyse results:
> "Seq Scan on assetregistericon  (cost=0.00..41106.50 rows=738050 width=428) (actual time=0.036..667.962 rows=738050 loops=1)"
> "Total runtime: 1209.181 ms"
> -- Executing query:
> select * from assetregistericon
> Total query runtime: 16345 ms.
> 738050 rows retrieved.
My first thought looking at the difference between the number given by
the analyse and the number given by (I presume) the client application
is that the transfer of the data between the server and the client
application is what's taking the time. Which would explain the near
constant total time for remote clients on different servers.

I presume there is some sort of id column on the table. What are the
execution times on the different servers for SELECT SUM(id) FROM
assetregistericon; ?

In response to

jnbpug by date

Next:From: renierDate: 2010-04-09 11:58:01
Subject: Re: PostgreSQL Tweaking
Previous:From: renierDate: 2010-04-09 10:57:11
Subject: Re: PostgreSQL Tweaking

Privacy Policy | About PostgreSQL
Copyright © 1996-2017 The PostgreSQL Global Development Group