Subject: Re: BUG #17321: count(*) on a 1,874,554,883 rows partitioned table
 takes several minutes.
To: deni_@hotmail.fr, pgsql-bugs@lists.postgresql.org
References: <17321-a10c625e59181269@postgresql.org>
From: Vik Fearing <vik@postgresfriends.org>
Message-ID: <93e50b9b-ffa4-2bfe-8b3c-5a286b9b2fc8@postgresfriends.org>
Date: Sat, 11 Dec 2021 04:14:33 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.14.0
MIME-Version: 1.0
In-Reply-To: <17321-a10c625e59181269@postgresql.org>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Archived-At: 
 <https://www.postgresql.org/message-id/93e50b9b-ffa4-2bfe-8b3c-5a286b9b2fc8%40postgresfriends.org>
Precedence: bulk

On 12/6/21 3:57 AM, PG Bug reporting form wrote:
> The following bug has been logged on the website:
> 
> Bug reference:      17321
> Logged by:          Denis Campredon
> Email address:      deni_@hotmail.fr
> PostgreSQL version: 13.3
> Operating system:   count(*) slow on partitioned table
> Description:        
> 
> I am auditing a database which has a really huge partitioned table.
> When working on the whole table queries are slower than the cumulated time
> for all partitions, as you can see with the following query. I don't have
> yet timings for more complex operations like group by, but it looks like it
> is slower than it should be.


If I'm reading this correctly, it counted 1579 rows /per millisecond/.
How may rows do you think it should have counted per millisecond?
-- 
Vik Fearing