<?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: Metadata Performance Exploration Part 2: XFS, JFS, ReiserFS, ext2, and Reiser4</title>
	<atom:link href="http://www.linux-mag.com/id/7518/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.linux-mag.com/id/7518/</link>
	<description>Open Source, Open Standards</description>
	<lastBuildDate>Sat, 05 Oct 2013 13:48:18 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.1</generator>
	<item>
		<title>By: jlauro</title>
		<link>http://www.linux-mag.com/id/7518/#comment-6959</link>
		<dc:creator>jlauro</dc:creator>
		<pubDate>Wed, 30 Nov -0001 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/7518/#comment-6959</guid>
		<description>&lt;p&gt;Great article with benchmark info.  It\&#039;s always good to have benchmarks completed on identical equipment.&lt;/p&gt;
&lt;p&gt;As a follow-up, it would be great to see how the \&quot;large\&quot; file systems compare.  Limit it to those that support at least 100TB file systems and 10TB files with 100,000 files in a single directory (or require even bigger limits).  With 2TB drives, 100TB file system is not as rare as it once was.&lt;/p&gt;
&lt;p&gt;Some of the options that claim to support that are unmanageable because a crash recovery is impractical because of memory requirements to complete.&lt;/p&gt;
&lt;p&gt;JFS is one option that scales well.  Most of the better performing options for the benchmark do not scale.  Not sure which other file systems do.
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>Great article with benchmark info.  It\&#8217;s always good to have benchmarks completed on identical equipment.</p>
<p>As a follow-up, it would be great to see how the \&#8221;large\&#8221; file systems compare.  Limit it to those that support at least 100TB file systems and 10TB files with 100,000 files in a single directory (or require even bigger limits).  With 2TB drives, 100TB file system is not as rare as it once was.</p>
<p>Some of the options that claim to support that are unmanageable because a crash recovery is impractical because of memory requirements to complete.</p>
<p>JFS is one option that scales well.  Most of the better performing options for the benchmark do not scale.  Not sure which other file systems do.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: gdrouillard</title>
		<link>http://www.linux-mag.com/id/7518/#comment-6960</link>
		<dc:creator>gdrouillard</dc:creator>
		<pubDate>Wed, 30 Nov -0001 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/7518/#comment-6960</guid>
		<description>&lt;p&gt;What mount options did you use in xfs and ext4?
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>What mount options did you use in xfs and ext4?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: laytonjb</title>
		<link>http://www.linux-mag.com/id/7518/#comment-6961</link>
		<dc:creator>laytonjb</dc:creator>
		<pubDate>Wed, 30 Nov -0001 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/7518/#comment-6961</guid>
		<description>&lt;p&gt;@gdrouillard:&lt;/p&gt;
&lt;p&gt;The xfs mount options were the defaults.&lt;br /&gt;
For ext4 I used the defaults except \&quot;data=ordered\&quot;.&lt;/p&gt;
&lt;p&gt;Jeff
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>@gdrouillard:</p>
<p>The xfs mount options were the defaults.<br />
For ext4 I used the defaults except \&#8221;data=ordered\&#8221;.</p>
<p>Jeff</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: laytonjb</title>
		<link>http://www.linux-mag.com/id/7518/#comment-6962</link>
		<dc:creator>laytonjb</dc:creator>
		<pubDate>Wed, 30 Nov -0001 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/7518/#comment-6962</guid>
		<description>&lt;p&gt;@jlauro,&lt;/p&gt;
&lt;p&gt;I would love to test those systems - do you have access to any hardware I can use for testing?&lt;/p&gt;
&lt;p&gt;Thanks!&lt;/p&gt;
&lt;p&gt;Jeff
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>@jlauro,</p>
<p>I would love to test those systems &#8211; do you have access to any hardware I can use for testing?</p>
<p>Thanks!</p>
<p>Jeff</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jlauro</title>
		<link>http://www.linux-mag.com/id/7518/#comment-6963</link>
		<dc:creator>jlauro</dc:creator>
		<pubDate>Wed, 30 Nov -0001 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/7518/#comment-6963</guid>
		<description>&lt;p&gt;@laytonjb: Unfortunately all the systems that I have access to over 10TB are tied up running production stuff...  I\&#039;ll try to keep this in mind next time we get a new SAN or new server for D2D backup.  A week or two of burn in running benchmarks is always a good idea, especially if it impacts how we leave the file system at for production...
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>@laytonjb: Unfortunately all the systems that I have access to over 10TB are tied up running production stuff&#8230;  I\&#8217;ll try to keep this in mind next time we get a new SAN or new server for D2D backup.  A week or two of burn in running benchmarks is always a good idea, especially if it impacts how we leave the file system at for production&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: porridge</title>
		<link>http://www.linux-mag.com/id/7518/#comment-6964</link>
		<dc:creator>porridge</dc:creator>
		<pubDate>Wed, 30 Nov -0001 00:00:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.linux-mag.com/id/7518/#comment-6964</guid>
		<description>&lt;p&gt;With this many numbers, it would be good to see them on a chart, rather than just in a table. Makes it much easier to compare.
&lt;/p&gt;
</description>
		<content:encoded><![CDATA[<p>With this many numbers, it would be good to see them on a chart, rather than just in a table. Makes it much easier to compare.</p>
]]></content:encoded>
	</item>
</channel>
</rss>