<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.0.11" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/">
<channel>
	<title>Comments on: MPI in Thirty Minutes</title>
	<link>http://www.linux-mag.com/id/5759/</link>
	<description>Open Source, Open Standards</description>
	<pubDate>Sun, 05 Jul 2009 01:45:28 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.11</generator>

	<item>
		<title>by: bejosukamto</title>
		<link>http://www.linux-mag.com/id/5759/#comment-2568</link>
		<pubDate>Thu, 07 May 2009 19:13:19 +0000</pubDate>
		<guid>http://www.linux-mag.com/id/5759/#comment-2568</guid>
					<description>Execuse Me....Anybody can help me, how to measure execution time for an elementary operations like addition, multiplication for processor AMD-Duron 800MHz, is't different with FLOPS? what the different way for FLOPS? thanks</description>
		<content:encoded><![CDATA[<p>Execuse Me&#8230;.Anybody can help me, how to measure execution time for an elementary operations like addition, multiplication for processor AMD-Duron 800MHz, is&#8217;t different with FLOPS? what the different way for FLOPS? thanks
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: zayar43</title>
		<link>http://www.linux-mag.com/id/5759/#comment-1756</link>
		<pubDate>Mon, 17 Nov 2008 19:18:34 +0000</pubDate>
		<guid>http://www.linux-mag.com/id/5759/#comment-1756</guid>
					<description>Great!!! Thanks!!!</description>
		<content:encoded><![CDATA[<p>Great!!! Thanks!!!
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: dave_lantos</title>
		<link>http://www.linux-mag.com/id/5759/#comment-1372</link>
		<pubDate>Thu, 07 Aug 2008 00:09:07 +0000</pubDate>
		<guid>http://www.linux-mag.com/id/5759/#comment-1372</guid>
					<description>What an informative, easy to follow, well written article.  Thanks for the excellent work, Joe!</description>
		<content:encoded><![CDATA[<p>What an informative, easy to follow, well written article.  Thanks for the excellent work, Joe!
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: drrossef</title>
		<link>http://www.linux-mag.com/id/5759/#comment-1238</link>
		<pubDate>Wed, 02 Jul 2008 10:09:49 +0000</pubDate>
		<guid>http://www.linux-mag.com/id/5759/#comment-1238</guid>
					<description>Nice intro - thanks</description>
		<content:encoded><![CDATA[<p>Nice intro - thanks
</p>
]]></content:encoded>
				</item>
	<item>
		<title>by: joe landman</title>
		<link>http://www.linux-mag.com/id/5759/#comment-1015</link>
		<pubDate>Sat, 19 Apr 2008 15:46:59 +0000</pubDate>
		<guid>http://www.linux-mag.com/id/5759/#comment-1015</guid>
					<description>Quick note:  the source can be downloaded from the above link, or from here http://downloads.scalableinformatics.com/downloads/linux-magazine/   
  ...  I mention this, as the OpenMP code link looks to be broken.  I'll let the editors know.

Also, there are a number of errors in this document (some formatting, some conversion).  The vector bit is supposed to be 

[ 1^(-2), 2^(-2), ... ]

or for the Fortran users among us

[ 1**(-2), 2**(-2), ...]

The idea is that we form a vector of integers from 1 to the "infinity" value, square each element, then invert each element.  This will give us a vector that looks like

[ 1, 1/4, 1/9, 1/16, .... ]

Then you can either sum these up, or notice that the definition of a dot product lets you do this as well

[ 1, 1/4, 1/9, 1/16, ...] dot [ 1, 1, 1, 1, ... ] = pi*pi/6

which was what I was trying to explain.  Once the formatting fell off, the rest didn't make much sense.

My apologies on this.</description>
		<content:encoded><![CDATA[<p>Quick note:  the source can be downloaded from the above link, or from here <a href="http://downloads.scalableinformatics.com/downloads/linux-magazine/" rel="nofollow">http://downloads.scalableinformatics.com/downloads/linux-magazine/</a><br />
  &#8230;  I mention this, as the OpenMP code link looks to be broken.  I&#8217;ll let the editors know.</p>
<p>Also, there are a number of errors in this document (some formatting, some conversion).  The vector bit is supposed to be </p>
<p>[ 1^(-2), 2^(-2), &#8230; ]</p>
<p>or for the Fortran users among us</p>
<p>[ 1**(-2), 2**(-2), &#8230;]</p>
<p>The idea is that we form a vector of integers from 1 to the &#8220;infinity&#8221; value, square each element, then invert each element.  This will give us a vector that looks like</p>
<p>[ 1, 1/4, 1/9, 1/16, &#8230;. ]</p>
<p>Then you can either sum these up, or notice that the definition of a dot product lets you do this as well</p>
<p>[ 1, 1/4, 1/9, 1/16, &#8230;] dot [ 1, 1, 1, 1, &#8230; ] = pi*pi/6</p>
<p>which was what I was trying to explain.  Once the formatting fell off, the rest didn&#8217;t make much sense.</p>
<p>My apologies on this.
</p>
]]></content:encoded>
				</item>
</channel>
</rss>
