Re: Delay locking partitions during query execution

From: Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>
To: David Rowley <david(dot)rowley(at)2ndquadrant(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Delay locking partitions during query execution
Date: 2019-01-03 13:40:39
Message-ID: 0d50a468-d261-e132-1516-bacbfdd844e7@2ndquadrant.com
Views: Raw Message | Whole Thread | Download mbox
Thread:
Lists: pgsql-hackers

On 12/3/18 12:42 PM, David Rowley wrote:
> ...
>
> Master: 10000 parts
>
> $ pgbench -n -f bench.sql -M prepared -T 60 postgres
> tps = 108.882749 (excluding connections establishing)
> tps = 108.245437 (excluding connections establishing)
>
> delaylock: 10000 parts
>
> $ pgbench -n -f bench.sql -M prepared -T 60 postgres
> tps = 1068.289505 (excluding connections establishing)
> tps = 1092.797015 (excluding connections establishing)
>

I'm a bit confused, because I can't reproduce any such speedup. I've
used the attached script that varies the number of partitions (which
worked quite nicely in the INSERT thread), but I'm getting results like
this:

partitions 0 100 1000 10000
--------------------------------------------
master 49 1214 186 11
patched 53 1225 187 11

So I don't see any significant speedup, for some reason :-(

Before I start digging into this, is there something that needs to be
done to enable it?

regards

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Attachment Content-Type Size
select.sql application/sql 60 bytes
run-select.sh application/x-shellscript 1002 bytes

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tomas Vondra 2019-01-03 13:51:43 Re: FETCH FIRST clause PERCENT option
Previous Message Peter Eisentraut 2019-01-03 13:28:51 Re: Unified logging system for command-line programs