Best practices for (plpgsql ?) trigger optimization? - Mailing list pgsql-general

From Karl O. Pinc
Subject Best practices for (plpgsql ?) trigger optimization?
Date
Msg-id 1112372395l.5596l.2l@mofo
Whole thread Raw
Responses Re: Best practices for (plpgsql ?) trigger optimization?  ("Karl O. Pinc" <kop@meme.com>)
List pgsql-general
Hi,

Are there any best practices for optimizing triggers,
and, I suppose, stored procedures as well?  I am now
starting on optimization and before I begin am
hoping to avoid re-inventing the wheel.


The problems I see are:

1) There is no way to profile where a problem lies.
When there are large and/or nested triggers there
could be a 'bad query' anywhere.  Finding it seems
difficult.

2) The NEW and OLD tables used by triggers don't exist
outside of a trigger environment, yet EXPLAIN returns
statement results -- and which is basically illegal inside
triggers.


The solutions I see are to use:

SET client_min_messages DEBUG1;
SET debug_print_plan TRUE;

and maybe
SET log_executer_stats TRUE;

Is this the best approach?  Any tricks for sorting through the
resultant output?

It'd be nice to have some hints about this in the User's Guide.

Thanks.

Karl <kop@meme.com>
Free Software:  "You don't pay back, you pay forward."
                 -- Robert A. Heinlein



pgsql-general by date:

Previous
From: Scott Marlowe
Date:
Subject: Re: importance of bgwriter_percent
Next
From: John Hughes
Date:
Subject: Inheritance and such