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;
	b=X8Zdj8uioGsT5ywFraocaY+TimabvU6Yb1E8BZ6TlEn73ai98qoXQs2U39QEtC+MYj
	MPEO+cPMlA5ClDJZCcJ268RR7xHSaVXpAhXENytEalD/kekzdI4WcRdik3MrRNpQ9fZW
	GPRGop8cA/UEk6AaM2DAKfkWOU2C9O8esnXRw=
MIME-Version: 1.0
In-Reply-To: <27894.1268836024@sss.pgh.pa.us>
References: <ca24673e1003161658g6a0fa30dg2c158db395ac08@mail.gmail.com>
	<20100317001405.GL3037@alvh.no-ip.org> <op.u9pa7vzleorkce@localhost>
	<407d949e1003170252y72d223c3kcc18266140a60997@mail.gmail.com>
	<27894.1268836024@sss.pgh.pa.us>
Date: Wed, 17 Mar 2010 10:16:00 -0500
Message-ID: <ca24673e1003170816r4690672xc83252fa3f71b23d@mail.gmail.com>
Subject: Re: Block at a time ...
From: Dave Crooke <dcrooke@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Greg Stark <gsstark@mit.edu>, Pierre C <lists@peufeu.com>,
	Alvaro Herrera <alvherre@commandprompt.com>,
	pgsql-performance@postgresql.org
Content-Type: multipart/alternative; boundary=001517588b34891a200482009798

--001517588b34891a200482009798
Content-Type: text/plain; charset=ISO-8859-1

Greg - with Oracle, I always do fixed 2GB dbf's for poartability, and
preallocate the whole file in advance. However, the situation is a bit
different in that Oracle will put blocks from multiple tables and indexes in
a DBF if you don't tell it differently.

Tom - I'm not sure what Oracle does, but it literally writes the whole
extent before using it .... I think they are just doing the literal
equivalent of *dd if=/dev/zero* ... it takes several seconds to prep a 2GB
file on decent storage.

Cheers
Dave

On Wed, Mar 17, 2010 at 9:27 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

> Greg Stark <gsstark@mit.edu> writes:
> > I think we need posix_fallocate().
>
> The problem with posix_fallocate (other than questionable portability)
> is that it doesn't appear to guarantee anything at all about what is in
> the space it allocates.  Worst case, we might find valid-looking
> Postgres data there (eg, because a block was recycled from some recently
> dropped table).  If we have to write something anyway to zero the space,
> what's the point?
>
>                        regards, tom lane
>

--001517588b34891a200482009798
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Greg - with Oracle, I always do fixed 2GB dbf&#39;s for poartability, and p=
reallocate the whole file in advance. However, the situation is a bit diffe=
rent in that Oracle will put blocks from multiple tables and indexes in a D=
BF if you don&#39;t tell it differently.<br>
<br>Tom - I&#39;m not sure what Oracle does, but it literally writes the wh=
ole extent before using it .... I think they are just doing the literal equ=
ivalent of <b>dd if=3D/dev/zero</b> ... it takes several seconds to prep a =
2GB file on decent storage.<br>
<br><div class=3D"gmail_quote">Cheers<br>Dave<br><br>On Wed, Mar 17, 2010 a=
t 9:27 AM, Tom Lane <span dir=3D"ltr">&lt;<a href=3D"mailto:tgl@sss.pgh.pa.=
us">tgl@sss.pgh.pa.us</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_q=
uote" style=3D"border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0=
pt 0.8ex; padding-left: 1ex;">
<div class=3D"im">Greg Stark &lt;<a href=3D"mailto:gsstark@mit.edu">gsstark=
@mit.edu</a>&gt; writes:<br>
&gt; I think we need posix_fallocate().<br>
<br>
</div>The problem with posix_fallocate (other than questionable portability=
)<br>
is that it doesn&#39;t appear to guarantee anything at all about what is in=
<br>
the space it allocates. =A0Worst case, we might find valid-looking<br>
Postgres data there (eg, because a block was recycled from some recently<br=
>
dropped table). =A0If we have to write something anyway to zero the space,<=
br>
what&#39;s the point?<br>
<br>
 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0regards, tom lane<br>
</blockquote></div><br>

--001517588b34891a200482009798--