<?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/"
	xmlns:series="http://unfoldingneurons.com/"
		>
<channel>
	<title>Comments on: Microsoft and Intel Pushing iSCSI Performance Limits</title>
	<atom:link href="http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/</link>
	<description>Understanding the accumulation of data</description>
	<lastBuildDate>Sat, 11 Feb 2012 20:58:00 -0500</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=</generator>
<atom:link rel="hub" href="http://pubsubhubbub.appspot.com" />
	<atom:link rel="hub" href="http://superfeedr.com/hubbub" />
		<item>
		<title>By: Microsoft and Intel Push One Million iSCSI IOPS &#8211; Gestalt IT</title>
		<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/comment-page-1/#comment-14208</link>
		<dc:creator>Microsoft and Intel Push One Million iSCSI IOPS &#8211; Gestalt IT</dc:creator>
		<pubDate>Fri, 15 Jan 2010 14:52:14 +0000</pubDate>
		<guid isPermaLink="false">http://blog.fosketts.net/?p=2590#comment-14208</guid>
		<description>[...] Microsoft and Intel Pushing iSCSI Performance Limits [...]</description>
		<content:encoded><![CDATA[<p>[...] Microsoft and Intel Pushing iSCSI Performance Limits [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: the storage anarchist</title>
		<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/comment-page-1/#comment-14170</link>
		<dc:creator>the storage anarchist</dc:creator>
		<pubDate>Fri, 08 Jan 2010 22:29:51 +0000</pubDate>
		<guid isPermaLink="false">http://blog.fosketts.net/?p=2590#comment-14170</guid>
		<description>We push such single-port rates out of both Symmetrix and CLARiiON. It&#039;s easiest to get this using multiple servers to generate the I/O load; I don&#039;t have the specifics about the HBAs/hosts used.&lt;br&gt;&lt;br&gt;II&#039;m not sure what the raw IOP numbers are for each platform...</description>
		<content:encoded><![CDATA[<p>We push such single-port rates out of both Symmetrix and CLARiiON. It&#39;s easiest to get this using multiple servers to generate the I/O load; I don&#39;t have the specifics about the HBAs/hosts used.</p>
<p>II&#39;m not sure what the raw IOP numbers are for each platform&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sfoskett</title>
		<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/comment-page-1/#comment-14169</link>
		<dc:creator>sfoskett</dc:creator>
		<pubDate>Fri, 08 Jan 2010 21:50:08 +0000</pubDate>
		<guid isPermaLink="false">http://blog.fosketts.net/?p=2590#comment-14169</guid>
		<description>What storage system pushed that kind of throughput over 8 Gb FC? And what HBA was used? That&#039;s much faster than anything I&#039;ve seen, and I&#039;m truly impressed that FC could deliver that kind of link efficiency!&lt;br&gt;&lt;br&gt;As for MS/Intel, I agree. Throughput is great, but what about latency? How many IOPS can be pushed through a 10 Gb iSCSI link? I can&#039;t wait to hear an answer to that question!&lt;br&gt;&lt;br&gt;For what it&#039;s worth, how many IOPS have you seen over an FC link?</description>
		<content:encoded><![CDATA[<p>What storage system pushed that kind of throughput over 8 Gb FC? And what HBA was used? That&#39;s much faster than anything I&#39;ve seen, and I&#39;m truly impressed that FC could deliver that kind of link efficiency!</p>
<p>As for MS/Intel, I agree. Throughput is great, but what about latency? How many IOPS can be pushed through a 10 Gb iSCSI link? I can&#39;t wait to hear an answer to that question!</p>
<p>For what it&#39;s worth, how many IOPS have you seen over an FC link?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: the storage anarchist</title>
		<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/comment-page-1/#comment-14168</link>
		<dc:creator>the storage anarchist</dc:creator>
		<pubDate>Fri, 08 Jan 2010 21:37:49 +0000</pubDate>
		<guid isPermaLink="false">http://blog.fosketts.net/?p=2590#comment-14168</guid>
		<description>I&#039;ve seen nearly .88 GigaBYTES/sec on 8Gb fibre as well.&lt;br&gt;&lt;br&gt;My point is that while it is commendable that one (apparently homogeneous) iSCSI driver stack is now able to attain &quot;link speed&quot; over 10GbE, that does not necessarily mean that FC cannot also deliver &quot;link speed.&quot; True, the max link speed of Ethernet is today faster than Fibre Channel, but the FC protocol is not necessarily any more limiting than iSCSI.&lt;br&gt;&lt;br&gt;And all that said, did MS/Intel report on small-block latencies? I suspect there may be more separation if the metric is small-block I/O (say, 512 byte reads/writes).</description>
		<content:encoded><![CDATA[<p>I&#39;ve seen nearly .88 GigaBYTES/sec on 8Gb fibre as well.</p>
<p>My point is that while it is commendable that one (apparently homogeneous) iSCSI driver stack is now able to attain &#8220;link speed&#8221; over 10GbE, that does not necessarily mean that FC cannot also deliver &#8220;link speed.&#8221; True, the max link speed of Ethernet is today faster than Fibre Channel, but the FC protocol is not necessarily any more limiting than iSCSI.</p>
<p>And all that said, did MS/Intel report on small-block latencies? I suspect there may be more separation if the metric is small-block I/O (say, 512 byte reads/writes).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sfoskett</title>
		<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/comment-page-1/#comment-14167</link>
		<dc:creator>sfoskett</dc:creator>
		<pubDate>Fri, 08 Jan 2010 20:47:51 +0000</pubDate>
		<guid isPermaLink="false">http://blog.fosketts.net/?p=2590#comment-14167</guid>
		<description>I&#039;ve seen greater than 1 gigabit per second on 4 Gb Fibre Channel links, too. But I&#039;ve never seen anything like the more than 1.1 gigaBYTES per second Microsoft and Intel showed in March. It&#039;s simply beyond the capabilities of any Fibre Channel SAN mathematically, let alone practically.&lt;br&gt;&lt;br&gt;How fast can the Symmetrix or Clariion push data out those iSCSI interfaces? I know they were recently upgraded from 1 Gb to 10 Gb, but how fast are they?</description>
		<content:encoded><![CDATA[<p>I&#39;ve seen greater than 1 gigabit per second on 4 Gb Fibre Channel links, too. But I&#39;ve never seen anything like the more than 1.1 gigaBYTES per second Microsoft and Intel showed in March. It&#39;s simply beyond the capabilities of any Fibre Channel SAN mathematically, let alone practically.</p>
<p>How fast can the Symmetrix or Clariion push data out those iSCSI interfaces? I know they were recently upgraded from 1 Gb to 10 Gb, but how fast are they?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: the storage anarchist</title>
		<link>http://blog.fosketts.net/2010/01/07/microsoft-intel-iscsi-performance/comment-page-1/#comment-14165</link>
		<dc:creator>the storage anarchist</dc:creator>
		<pubDate>Fri, 08 Jan 2010 17:08:51 +0000</pubDate>
		<guid isPermaLink="false">http://blog.fosketts.net/?p=2590#comment-14165</guid>
		<description>For the record, both Symm and CLARiiON routinely deliver &gt; 1Gbps over single FC links. Server-side configuration is usually the bottleneck (queue depth, etc.)...iSCSI can be limited by these same settings - would be great to see apples-apples comparisons with equivalent configurations.&lt;br&gt;&lt;br&gt;Oh, and Symmetrix is still the only enterprise-class storage array with native iSCSI support. It&#039;s great to see the performance advances being made by Microsoft and Intel- hopefully will drive up adoption across the board.</description>
		<content:encoded><![CDATA[<p>For the record, both Symm and CLARiiON routinely deliver &gt; 1Gbps over single FC links. Server-side configuration is usually the bottleneck (queue depth, etc.)&#8230;iSCSI can be limited by these same settings &#8211; would be great to see apples-apples comparisons with equivalent configurations.</p>
<p>Oh, and Symmetrix is still the only enterprise-class storage array with native iSCSI support. It&#39;s great to see the performance advances being made by Microsoft and Intel- hopefully will drive up adoption across the board.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

