Message-ID: <50FFF659.6070003@gmail.com>
Date: Wed, 23 Jan 2013 22:40:25 +0800
From: Rural Hunter <ruralhunter@gmail.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1;
	rv:17.0) Gecko/20130107 Thunderbird/17.0.2
MIME-Version: 1.0
To: pgsql-admin@postgresql.org
Subject: Re: vacuum / analyze parent tables on partitioned tables.
References: 
 <CAFCtE1k+8iRr2OnSRr6vaRC7+ArqMShZ3Tnd_H1tqV-ahhpsng@mail.gmail.com>
In-Reply-To: 
 <CAFCtE1k+8iRr2OnSRr6vaRC7+ArqMShZ3Tnd_H1tqV-ahhpsng@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Sender: pgsql-admin-owner@postgresql.org

I'm doing the same thing. In my case, the vacuum part on parent is very 
quick while analyzing takes a bit longer since it runs rough analyzes 
all children tables. You can see the behavior by "analyze verbose". 
Maybe the bigger part of your vacuum/analyze is on analyze so that you 
are seeing this result.

于 2013/1/23 19:43, Bert 写道:
> Hello,
>
> I first wrote, by mistake, to the sql mailing list. But here is my 
> e-mail:
>
> I wrote a script to make sure all tables are vacuumed and analyzed 
> every evening. This works very well.
> I save in a table the start and end time of a vacuum/analyze. This way 
> I can measure what tables take a long time to vaccum/analyze, and what 
> tables are slow. (and much more).
>
> But I have noticed that the parent table of a partitioned table also 
> takes a long time. Here is a snap shot of the following table
>
> table_name          ;     avg runt time   ;   max run time  ;   min 
> run time
> "f_transaction_1"  ;    "00:03:07.8"     ;   "00:03:10" ;   "00:03:03"
> "f_transaction"      ;   "00:02:19.8"     ;   "00:02:25" ;  "00:02:16"
>
> f_tranaction_1 is 16GB data + 12GB of indexes. (I know, a lot of 
> indexes). f_tranaction is totally empy, but also contains all indexes. 
> Which means 0B table zise, and 140kB index size.
>
> Does anyone has an idea why in this case the vacuum/analyze takes 
> almost as long on the parent table as on the biggest child table? (the 
> other child tables are smaller than f_tranaction_1, and their 
> vacuum/analyze time is much shorter).
>
> wkr,
> Bert


-- 
Sent via pgsql-admin mailing list (pgsql-admin@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin