Re: query planning different in plpgsql? - Mailing list pgsql-performance

From Michal J. Kubski
Subject Re: query planning different in plpgsql?
Date
Msg-id 0ddf8c22989a34dc924fef1897bdc8d7@localhost
Whole thread Raw
In response to Re: query planning different in plpgsql?  (Grzegorz Jaśkiewicz <gryzman@gmail.com>)
Responses Re: query planning different in plpgsql?  (Merlin Moncure <mmoncure@gmail.com>)
List pgsql-performance

Hi,

On Fri, 23 Oct 2009 16:56:36 +0100, Grzegorz Jaśkiewicz
<gryzman@gmail.com> wrote:
> On Fri, Oct 23, 2009 at 4:49 PM, Scott Mead
> <scott.lists@enterprisedb.com>wrote:
> 
>>
>>
>>   Do you not have an index on last_snapshot.domain_id?
>>
> 
> that, and also try rewriting a query as JOIN. There might be difference
in
> performance/plan.
> 
Thanks, it runs better (average 240s, not 700s) with the index. Rewriting
queries
as JOINs does not make any difference.
The last_snapshot is a temp table created earlier in the procedure
and the query in question is preceded with CREATE TEMPORARY TABLE as well,
not a cursor. 
I still do not get why it performs differently inside the procedure. 
Is there any way to see what planning decisions were made?

Best regards,
Michal

-- 
I hear and I forget. I see and I believe. I do and I understand.
(Confucius)


pgsql-performance by date:

Previous
From: "Marc Mamin"
Date:
Subject: Re: optimizing query with multiple aggregates
Next
From: Merlin Moncure
Date:
Subject: Re: query planning different in plpgsql?