<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Linux Scalability in a NUMA World</title>
	<atom:link href="http://www.linux-mag.com/id/6868/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.linux-mag.com/id/6868/</link>
	<description>Open Source, Open Standards</description>
	<lastBuildDate>Fri, 10 Feb 2012 10:47:01 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.1</generator>
	<item>
		<title>By: cavemole</title>
		<link>http://www.linux-mag.com/id/6868/#comment-5681</link>
		<dc:creator>cavemole</dc:creator>
		<pubDate>Tue, 30 Nov 1999 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/6868/#comment-5681</guid>
		<description>Excellent introductory article.&lt;br /&gt;
I will be keeping this in mind while designing a new application.&lt;br /&gt;
I always wondered if there was a way to control aspects of NUMA.&lt;br /&gt;
I now have a starting point for researching these NUMA control techniques.</description>
		<content:encoded><![CDATA[<p>Excellent introductory article.<br />
I will be keeping this in mind while designing a new application.<br />
I always wondered if there was a way to control aspects of NUMA.<br />
I now have a starting point for researching these NUMA control techniques.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: dcymbala</title>
		<link>http://www.linux-mag.com/id/6868/#comment-5682</link>
		<dc:creator>dcymbala</dc:creator>
		<pubDate>Tue, 30 Nov 1999 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/6868/#comment-5682</guid>
		<description>I&#039;m curious if there is any support in the Linux kernel for an application to be profiled so that the NUMA model it uses is based on some kind of empirical data. It seems that right now it is up to the application writer to have some understanding of the memory usage behavior the application has in advance. I suppose that each application could provide command-line parameters to select a model, but it would be really great for the application to &quot;observe itself&quot; use memory and adapt dynamically. Do such things exist right now?</description>
		<content:encoded><![CDATA[<p>I&#8217;m curious if there is any support in the Linux kernel for an application to be profiled so that the NUMA model it uses is based on some kind of empirical data. It seems that right now it is up to the application writer to have some understanding of the memory usage behavior the application has in advance. I suppose that each application could provide command-line parameters to select a model, but it would be really great for the application to &#8220;observe itself&#8221; use memory and adapt dynamically. Do such things exist right now?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

