X-Original-To: pgsql-hackers-postgresql.org@localhost.postgresql.org
Received: from localhost (av.hub.org [200.46.204.144])
	by postgresql.org (Postfix) with ESMTP id D50209DCC37
	for <pgsql-hackers-postgresql.org@localhost.postgresql.org>;
	Tue, 29 Nov 2005 01:21:01 -0400 (AST)
Received: from postgresql.org ([200.46.204.71])
	by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024)
	with ESMTP id 09942-09
	for <pgsql-hackers-postgresql.org@localhost.postgresql.org>;
	Tue, 29 Nov 2005 01:21:00 -0400 (AST)
X-Greylist: from auto-whitelisted by SQLgrey-
Received: from toad.mtbrook.bozemanpass.com (toad.mtbrook.bozemanpass.com
	[69.145.82.195])
	by postgresql.org (Postfix) with ESMTP id D3D3C9DCAB3
	for <pgsql-hackers@postgresql.org>;
	Tue, 29 Nov 2005 01:20:58 -0400 (AST)
Received: from [69.145.82.253] (unknown [69.145.82.253])
	by toad.mtbrook.bozemanpass.com (Postfix) with ESMTP id 915B311027B
	for <pgsql-hackers@postgresql.org>;
	Mon, 28 Nov 2005 21:20:58 -0800 (PST)
Message-ID: <438BE53B.1090106@boreham.org>
Date: Mon, 28 Nov 2005 22:20:59 -0700
From: David Boreham <david_list@boreham.org>
User-Agent: Mozilla Thunderbird 1.0.7 (Windows/20050923)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: pgsql-hackers@postgresql.org
Subject: Re: ice-broker scan thread
References: <Pine.LNX.4.58.0511282217470.13586@josh.db>
	<Pine.LNX.4.58.0511291429330.18112@linuxworld.com.au>
	<Pine.LNX.4.58.0511282350370.13833@josh.db>
In-Reply-To: <Pine.LNX.4.58.0511282350370.13833@josh.db>
Content-Type: multipart/alternative;
	boundary="------------050800010703040302050000"
X-Virus-Scanned: by amavisd-new at hub.org
X-Spam-Status: No, score=0.001 required=5 tests=[AWL=-0.000,
	HTML_MESSAGE=0.001]
X-Spam-Score: 0.001
X-Spam-Level: 
X-Archive-Number: 200511/1463
X-Sequence-Number: 76745

This is a multi-part message in MIME format.
--------------050800010703040302050000
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Qingqing Zhou wrote:

>On Mon, 28 Nov 2005, Gavin Sherry wrote:
>  
>
>>MySQL, Oracle and others implement read-ahead threads to simulate async IO
>>'pre-fetching'.
>>    
>>
>
>Due to my tests on Windows (using the attached program and change
>enable_aio=true), seems aio doesn't help as a separate thread - but maybe
>because my usage is wrong ...
>  
>
I don't think your NT overlapped I/O code is quite right. At least
I think it will issue reads at a high rate without waiting for any of them
to complete. Beyond some point that has to give the kernel gut-rot.
But anyway, I wouldn't expect the use of aio to make any
significant difference in an already threaded test program.
The point of aio is to allow
I/O concurrency _without_ the use of threads or multiple processes.
You could re-write your program to have a single thread but use aio.
In that case it should show the same read ahead benefit that you see
with the thread.




--------------050800010703040302050000
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Qingqing Zhou wrote:
<blockquote cite="midPine.LNX.4.58.0511282350370.13833@josh.db"
 type="cite">
  <pre wrap="">
On Mon, 28 Nov 2005, Gavin Sherry wrote:
  </pre>
  <blockquote type="cite">
    <pre wrap="">MySQL, Oracle and others implement read-ahead threads to simulate async IO
'pre-fetching'.
    </pre>
  </blockquote>
  <pre wrap=""><!---->
Due to my tests on Windows (using the attached program and change
enable_aio=true), seems aio doesn't help as a separate thread - but maybe
because my usage is wrong ...
  </pre>
</blockquote>
I don't think your NT overlapped I/O code is quite right. At least<br>
I think it will issue reads at a high rate without waiting for any of
them<br>
to complete. Beyond some point that has to give the kernel gut-rot.<br>
But anyway, I wouldn't expect the use of aio to make any<br>
significant difference in an already threaded test program. <br>
The point of aio is to allow<br>
I/O concurrency _without_ the use of threads or multiple processes.<br>
You could re-write your program to have a single thread but use aio.<br>
In that case it should show the same read ahead benefit that you see<br>
with the thread.<br>
<br>
<br>
<br>
</body>
</html>

--------------050800010703040302050000--