Date: Fri, 2 Nov 2018 15:51:40 -0700 (MST)
From: legrand legrand <legrand_legrand@hotmail.com>
To: pgsql-performance@postgresql.org
Message-ID: <1541199100540-0.post@n3.nabble.com>
In-Reply-To: <8db7abc2-9ddf-858c-e3db-af014cf2ccd9@2ndquadrant.com>
References: 
 <CALhvPTS3XmDa7Px2WdZhyvwHPsqC=KKQ2kTsGL0KZiuwGF8C4g@mail.gmail.com>
 <8db7abc2-9ddf-858c-e3db-af014cf2ccd9@2ndquadrant.com>
Subject: Re: Fwd: Query with high planning time compared to execution time
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Precedence: bulk

Tomas Vondra-4 wrote
> On 11/02/2018 10:36 AM, Richard Lee wrote:
> [...]
> 
>> What are my other options to improve the query planning time?
>> 
> 
> Can you do a bit of profiling, to determine which part of the query
> planning process is slow here? 
> [...]

After planning profiling, (or in //), you can try to limit the number of
plans 
that the planner has to evaluate:

setting enable_mergejoin to off, or some others from 
https://www.postgresql.org/docs/10/static/runtime-config-query.html
but you will have to check that the chosen plan is still good

An other way is maybe reducing the number of indexes (you have so many 
ones ...). Usually, needed indexes are PK, UK, indexes for FK, and a "few" 
more.

Could you provide the SQL query to check that ?

Regards
PAscal  


--
Sent from: http://www.postgresql-archive.org/PostgreSQL-performance-f2050081.html