pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Anibal David Acosta <aa@devshock.com>
To: tv@fuzzy.cz
Cc: pgsql-performance@postgresql.org
Subject: Re: how much postgres can scale up?
Date: Fri, 10 Jun 2011 08:56:50 -0400
Message-ID: <000f01cc276d$df82edc0$9e88c940$@devshock.com> (raw)
In-Reply-To: <243bcc276983a6b1e2bbc783c3a5b7db.squirrel@sq.gransy.com>
References: <000601cc2761$a2e3a280$e8aae780$@devshock.com>
	<243bcc276983a6b1e2bbc783c3a5b7db.squirrel@sq.gransy.com>

The version is Postgres 9.0
Yes, I setup the postgres.conf according to instructions in the 
http://wiki.postgresql.org/wiki/Tuning_Your_PostgreSQL_Server


Cool, I will check this
http://wiki.postgresql.org/wiki/Logging_Difficult_Queries

Looks like great starting point to find bottleneck

But so, Is possible in excellent conditions that two connections duplicate the quantity of transactions per second?

Thanks!


-----Mensaje original-----
De: tv@fuzzy.cz [mailto:tv@fuzzy.cz] 
Enviado el: viernes, 10 de junio de 2011 08:10 a.m.
Para: Anibal David Acosta
CC: pgsql-performance@postgresql.org
Asunto: Re: [PERFORM] how much postgres can scale up?

> I have a function in pgsql language, this function do some select to 
> some tables for verify some conditions and then do one insert to a 
> table with NO index. Update are not performed in the function
>
> When 1 client connected postgres do 180 execution per second With 2 
> clients connected postgres do 110 execution per second With 3 clients 
> connected postgres do 90 execution per second
>
> Finally with 6 connected clients postgres do 60 executions per second 
> (totally 360 executions per second)
>
> While testing, I monitor disk, memory and CPU and not found any overload.

There's always a bottleneck - otherwise the system might run faster (and hit another bottleneck eventually). It might be CPU, I/O, memory, locking and maybe some less frequent things.

> I know that with this information you can figure out somethigns, but 
> in normal conditions, Is normal the degradation of performance per 
> connection when connections are incremented?
> Or should I spect 180 in the first and something similar in the second 
> connection? Maybe 170?
>
>
> The server is a dual xeon quad core with 16 GB of ram and a very fast 
> storage The OS is a windows 2008 R2 x64

Might be, but we need more details about how the system works. On Linux I'd ask for output from 'iostat -x 1' and 'vmstat 1' but you're on Windows so there are probably other tools.

What version of PostgreSQL is this? What are the basic config values (shared_buffers, work_mem, effective_cache_size, ...)? Have you done some tuning? There's a wiki page about this:
http://wiki.postgresql.org/wiki/Tuning_Your_PostgreSQL_Server

Have you tried to log slow queries? Maybe there's one query that makes the whole workload slow? See this:
http://wiki.postgresql.org/wiki/Logging_Difficult_Queries

Tomas




view thread (11+ messages)  latest in thread

Message-ID: <000f01cc276d$df82edc0$9e88c940$@devshock.com>
Permalink:  ../000f01cc276d$df82edc0$9e88c940$@devshock.com/
Also on:    postgresql.org/message-id/000f01cc276d$df82edc0$9e88c940$@devshock.com

 · 

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-performance@postgresql.org
  Cc: aa@devshock.com, tv@fuzzy.cz
  Subject: Re: how much postgres can scale up?
  In-Reply-To: <000f01cc276d$df82edc0$9e88c940$@devshock.com>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox