<?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: Using Solid State Disks on Linux</title>
	<atom:link href="http://danweinreb.org/blog/using-solid-state-disks-on-linux/feed" rel="self" type="application/rss+xml" />
	<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux</link>
	<description>Software and Innovation</description>
	<lastBuildDate>Wed, 01 Sep 2010 14:32:43 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
	<item>
		<title>By: Dan Weinreb</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-48244</link>
		<dc:creator>Dan Weinreb</dc:creator>
		<pubDate>Wed, 28 Jul 2010 14:03:17 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-48244</guid>
		<description>Re the last comment, the referencing page has some useful practical advice.  It&#039;s worth reading:

http://optimisticpanda.co.uk/blog/?p=1</description>
		<content:encoded><![CDATA[<p>Re the last comment, the referencing page has some useful practical advice.  It&#8217;s worth reading:</p>
<p><a href="http://optimisticpanda.co.uk/blog/?p=1" rel="nofollow">http://optimisticpanda.co.uk/blog/?p=1</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Linux With SSDs &#171; optimisticpanda.co.uk</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-48237</link>
		<dc:creator>Linux With SSDs &#171; optimisticpanda.co.uk</dc:creator>
		<pubDate>Tue, 27 Jul 2010 18:31:51 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-48237</guid>
		<description>[...] taken from http://danweinreb.org/blog/using-solid-state-disks-on-linux [...]</description>
		<content:encoded><![CDATA[<p>[...] taken from <a href="http://danweinreb.org/blog/using-solid-state-disks-on-linux" rel="nofollow">http://danweinreb.org/blog/using-solid-state-disks-on-linux</a> [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: How to Get Six Pack Fast</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-24264</link>
		<dc:creator>How to Get Six Pack Fast</dc:creator>
		<pubDate>Wed, 15 Apr 2009 16:27:30 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-24264</guid>
		<description>If you   want to read a reader&#039;s feedback :) , I rate this post for 4/5. Detailed info, but I just have to go to that damn google to find the missed bits. Thank you, anyway!</description>
		<content:encoded><![CDATA[<p>If you   want to read a reader&#8217;s feedback <img src='http://danweinreb.org/blog/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' />  , I rate this post for 4/5. Detailed info, but I just have to go to that damn google to find the missed bits. Thank you, anyway!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: anon</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-13334</link>
		<dc:creator>anon</dc:creator>
		<pubDate>Sun, 22 Feb 2009 22:54:48 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-13334</guid>
		<description>After fearther reading (particularly this thread http://lkml.org/lkml/2008/10/14/129 ) to try to respond to my precedent questions, here&#039;s the big picture I&#039;ve found about the current commercial wave of commonly available SSDs:

 - There is no documented access to the raw NAND hardware, so you can not effectively use well known dedicated FS like UBIFS, JFFS2, 
 -  Manufacturers implement a proprietary hardware layer on top of the raw flash and you are stuck with that. Actually, there is no standardized good controller technology and that is probably the origin of the confusion on the file system side. 
 - For reliability and performance you need to pay many bucks, and with that hardware, every common file system (ext3 etc) perform well.

Other interesting readings I&#039;ve found on this topic :
http://valhenson.livejournal.com/25228.html
http://realworldtech.com/forums/index.cfm?action=detail&amp;id=92756&amp;threadid=92678&amp;roomid=2</description>
		<content:encoded><![CDATA[<p>After fearther reading (particularly this thread <a href="http://lkml.org/lkml/2008/10/14/129" rel="nofollow">http://lkml.org/lkml/2008/10/14/129</a> ) to try to respond to my precedent questions, here&#8217;s the big picture I&#8217;ve found about the current commercial wave of commonly available SSDs:</p>
<p> &#8211; There is no documented access to the raw NAND hardware, so you can not effectively use well known dedicated FS like UBIFS, JFFS2,<br />
 &#8211;  Manufacturers implement a proprietary hardware layer on top of the raw flash and you are stuck with that. Actually, there is no standardized good controller technology and that is probably the origin of the confusion on the file system side.<br />
 &#8211; For reliability and performance you need to pay many bucks, and with that hardware, every common file system (ext3 etc) perform well.</p>
<p>Other interesting readings I&#8217;ve found on this topic :<br />
<a href="http://valhenson.livejournal.com/25228.html" rel="nofollow">http://valhenson.livejournal.com/25228.html</a><br />
<a href="http://realworldtech.com/forums/index.cfm?action=detail&amp;id=92756&amp;threadid=92678&amp;roomid=2" rel="nofollow">http://realworldtech.com/forums/index.cfm?action=detail&amp;id=92756&amp;threadid=92678&amp;roomid=2</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: anon</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-13199</link>
		<dc:creator>anon</dc:creator>
		<pubDate>Sun, 22 Feb 2009 14:14:27 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-13199</guid>
		<description>Does you really need a special file system to use a modern SSD that implement in hardware a &#039;wear levelling&#039; algorithm ?

JFFS2 (or a more modern flash file system like UBIFS ) is somewhat necessary on flash devices without hardware to perform wear levelling. 

It is clear that those new file systems carries in general better technology for flash devices that traditional file systems which were developed for spinning HDDs, But how do they deal with the fact that modern SSDs manufacturers include a hardware layer in their devices to let them perform well under a traditional OS ?

UBIFS :
http://www.inf.u-szeged.hu/sed/ubifs
http://lwn.net/Articles/276025/</description>
		<content:encoded><![CDATA[<p>Does you really need a special file system to use a modern SSD that implement in hardware a &#8216;wear levelling&#8217; algorithm ?</p>
<p>JFFS2 (or a more modern flash file system like UBIFS ) is somewhat necessary on flash devices without hardware to perform wear levelling. </p>
<p>It is clear that those new file systems carries in general better technology for flash devices that traditional file systems which were developed for spinning HDDs, But how do they deal with the fact that modern SSDs manufacturers include a hardware layer in their devices to let them perform well under a traditional OS ?</p>
<p>UBIFS :<br />
<a href="http://www.inf.u-szeged.hu/sed/ubifs" rel="nofollow">http://www.inf.u-szeged.hu/sed/ubifs</a><br />
<a href="http://lwn.net/Articles/276025/" rel="nofollow">http://lwn.net/Articles/276025/</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dan Weinreb</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-13171</link>
		<dc:creator>Dan Weinreb</dc:creator>
		<pubDate>Sun, 22 Feb 2009 12:06:33 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-13171</guid>
		<description>Andrew: Thank you!  Fixed.

Nat: I don&#039;t know.  I saw at least two recommendations to make this setting, though.

A friend has been telling me that I shouldn&#039;t even think of recommending ext2 over ext3.  Of course, I totally agree that logging the metadata is a great idea.  I was more trying to express the fact that there is so little advice out there about what Linux file system to use, and what advice you do find is not always terribly convincing.  I was very surprised by this.  I thought I&#039;d do a quick Google search, and find out that &quot;everybody knows&quot; that the best Linux file system for SSD&#039;s is such-and-such, and that would be it.  I still don&#039;t know why it didn&#039;t turn out that way.</description>
		<content:encoded><![CDATA[<p>Andrew: Thank you!  Fixed.</p>
<p>Nat: I don&#8217;t know.  I saw at least two recommendations to make this setting, though.</p>
<p>A friend has been telling me that I shouldn&#8217;t even think of recommending ext2 over ext3.  Of course, I totally agree that logging the metadata is a great idea.  I was more trying to express the fact that there is so little advice out there about what Linux file system to use, and what advice you do find is not always terribly convincing.  I was very surprised by this.  I thought I&#8217;d do a quick Google search, and find out that &#8220;everybody knows&#8221; that the best Linux file system for SSD&#8217;s is such-and-such, and that would be it.  I still don&#8217;t know why it didn&#8217;t turn out that way.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Andrew Benton</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-13042</link>
		<dc:creator>Andrew Benton</dc:creator>
		<pubDate>Sun, 22 Feb 2009 03:08:38 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-13042</guid>
		<description>You missed an r of the end of the command

cat /sys/block/sda/queue/scheduler</description>
		<content:encoded><![CDATA[<p>You missed an r of the end of the command</p>
<p>cat /sys/block/sda/queue/scheduler</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nat Tuck</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-13032</link>
		<dc:creator>Nat Tuck</dc:creator>
		<pubDate>Sun, 22 Feb 2009 02:35:00 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-13032</guid>
		<description>How many writes does noatime save compared to relatime?</description>
		<content:encoded><![CDATA[<p>How many writes does noatime save compared to relatime?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Fred Blasdel</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-12947</link>
		<dc:creator>Fred Blasdel</dc:creator>
		<pubDate>Sat, 21 Feb 2009 20:48:40 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-12947</guid>
		<description>JFFS2 is designed for the raw MTD flash that&#039;s been used in embedded devices for a long time, where the device does no wear-leveling or write-batching, it&#039;s as dumb as it can be.

JFFS is not for generic block devices that happen to be flash underneath. It&#039;s already slow enough on MTDs, using it pointlessly on a SSD would have pathological performance.</description>
		<content:encoded><![CDATA[<p>JFFS2 is designed for the raw MTD flash that&#8217;s been used in embedded devices for a long time, where the device does no wear-leveling or write-batching, it&#8217;s as dumb as it can be.</p>
<p>JFFS is not for generic block devices that happen to be flash underneath. It&#8217;s already slow enough on MTDs, using it pointlessly on a SSD would have pathological performance.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: anon</title>
		<link>http://danweinreb.org/blog/using-solid-state-disks-on-linux/comment-page-1#comment-12945</link>
		<dc:creator>anon</dc:creator>
		<pubDate>Sat, 21 Feb 2009 20:42:17 +0000</pubDate>
		<guid isPermaLink="false">http://danweinreb.org/blog/?p=119#comment-12945</guid>
		<description>(oops. sorry. read too quickly and missed you cited his article already.)</description>
		<content:encoded><![CDATA[<p>(oops. sorry. read too quickly and missed you cited his article already.)</p>
]]></content:encoded>
	</item>
</channel>
</rss>
