DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma;
	h=mime-version:in-reply-to:references:date:message-id:subject:from:to
	:cc:content-type:content-transfer-encoding;
	b=UTZAVUweRnqDxsU6OngxgZPpD/DTkmr0fkxTUD3GgNsQyEtKSXF+xT9XEIEzTgvr0S
	8BltutIOJ2IeFthtFOtfeTq/kvFAjWydmbuTkc5DLyKNDdCc2yVMshcuBoT4cHKbHoOu
	LOMDHpNjmDfR6kVQNpnladlS4/G6qQvy6jZ+E=
MIME-Version: 1.0
In-Reply-To: <20100107141423.GC4315@alvh.no-ip.org>
References: <852badbc1001070438j36eb10d3qa22398ffcbef405f@mail.gmail.com>
	<4B45E36E.4040706@tweakers.net>
	<852badbc1001070547p48b47175vf87efc473e3bd41b@mail.gmail.com>
	<26014999.20100107150535@erwied.eu>
	<852badbc1001070610p698dc3cfr4628b1e7768990e6@mail.gmail.com>
	<20100107141423.GC4315@alvh.no-ip.org>
Date: Thu, 7 Jan 2010 15:23:24 +0100
Message-ID: <852badbc1001070623m5d901c42k50bb808409141ca4@mail.gmail.com>
Subject: Re: Air-traffic benchmark
From: Lefteris <lsidir@gmail.com>
To: Alvaro Herrera <alvherre@commandprompt.com>
Cc: Jochen Erwied <jochen@pgsql-performance.erwied.eu>,
	pgsql-performance@postgresql.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Thu, Jan 7, 2010 at 3:14 PM, Alvaro Herrera
<alvherre@commandprompt.com> wrote:
> Lefteris escribi=F3:
>> Yes, I am reading the plan wrong! I thought that each row from the
>> plan reported the total time for the operation but it actually reports
>> the starting and ending point.
>>
>> So we all agree that the problem is on the scans:)
>>
>> So the next question is why changing shared memory buffers will fix
>> that? i only have one session with one connection, do I have like many
>> reader workers or something?
>
> No amount of tinkering is going to change the fact that a seqscan is the
> fastest way to execute these queries. =A0Even if you got it to be all in
> memory, it would still be much slower than the other systems which, I
> gather, are using columnar storage and thus are perfectly suited to this
> problem (unlike Postgres). =A0The talk about "compression ratios" caught
> me by surprise until I realized it was columnar stuff. =A0There's no way
> you can get such high ratios on a regular, row-oriented storage.
>
> --
> Alvaro Herrera =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0http://www.CommandPrompt.com/
> PostgreSQL Replication, Consulting, Custom Development, 24x7 support
>

I am aware of that and I totally agree. I would not expect from a row
store to have the same performance with a column. I was just trying to
double check that all settings are correct because usually you have
difference of seconds and minutes between column-rows, not seconds and
almost an hour (for queries Q2-Q8).

I think what you all said was very helpful and clear! The only part
that I still disagree/don't understand is the shared_buffer option:))

Lefteris