<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	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:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>Professional Geekism &#187; infiniband</title>
	<atom:link href="http://www.ninjabadger.net/tag/infiniband/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.ninjabadger.net</link>
	<description>Ninjas. Badgers. Linux. Me.</description>
	<lastBuildDate>Fri, 05 Aug 2011 13:49:31 +0000</lastBuildDate>
	<language>en</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.2.1</generator>
		<item>
		<title>Crossing the Gigabit barrier</title>
		<link>http://www.ninjabadger.net/2008/04/28/crossing-the-gigabit-barrier/</link>
		<comments>http://www.ninjabadger.net/2008/04/28/crossing-the-gigabit-barrier/#comments</comments>
		<pubDate>Mon, 28 Apr 2008 10:12:45 +0000</pubDate>
		<dc:creator>Tom</dc:creator>
				<category><![CDATA[Networking]]></category>
		<category><![CDATA[Storage]]></category>
		<category><![CDATA[10gigE]]></category>
		<category><![CDATA[fibrechannel]]></category>
		<category><![CDATA[infiniband]]></category>
		<category><![CDATA[nas]]></category>
		<category><![CDATA[nfs]]></category>
		<category><![CDATA[san]]></category>
		<category><![CDATA[sun]]></category>
		<category><![CDATA[ZFS]]></category>

		<guid isPermaLink="false">http://www.ninjabadger.net/?p=27</guid>
		<description><![CDATA[Recently, I&#8217;ve been charged with investigating into faster-than-gigabit networking, in an effort to switch our VM hosts away from local storage to an NFS-based NAS system. There are a few reasons for doing this; the greatest of which is Sun&#8217;s ZFS file system. ZFS, for those of you who aren&#8217;t familiar, has really shaken-up the [...]]]></description>
			<content:encoded><![CDATA[<p>Recently, I&#8217;ve been charged with investigating into faster-than-gigabit networking, in an effort to switch our VM hosts away from local storage to an <a href="http://en.wikipedia.org/wiki/Network_File_System_(protocol)">NFS</a>-based <a href="http://en.wikipedia.org/wiki/Network_Attached_Storage">NAS</a> system. There are a few reasons for doing this; the greatest of which is Sun&#8217;s <a href="http://en.wikipedia.org/wiki/ZFS">ZFS</a> file system.</p>
<p>ZFS, for those of you who aren&#8217;t familiar, has really shaken-up the world of file systems recently, as it changes almost everything that we perceive about a modern-day file system. On top of these fundamental changes (which I won&#8217;t go into detail about here) the ZFS developers have added some really neat features, such-as zero-cost snaphosts, replication between machines, RAID-Z, and quite a lot more.</p>
<p>It&#8217;s the promise of these features that has prompted our change over to a NAS-based storage system. Given that we can completely replace our current system of identical live/backup hosts, with slow backup scripts and <a href="http://en.wikipedia.org/wiki/Drbd">drbd</a> mirroring, it&#8217;s quite promising to think what we can achieve.</p>
<p>The problem is transport. And keeping fast transport. Given the extra overheads of IP/NFS that NAS brings (weighted against the benefits given ZFS over the more efficient use of raw disks in a <a href="http://en.wikipedia.org/wiki/Storage_Area_Network_(SAN)">SAN</a>) it&#8217;s been deemed that a single gigabit link just won&#8217;t be up to the demanding task. The problem is that once you decide to cross the gigabit &#8216;barrier&#8217;, your costing simply spirals uncontrollably skyward. <img src='http://www.ninjabadger.net/wp-includes/images/smilies/icon_sad.gif' alt=':(' class='wp-smiley' /> </p>
<p>There are a few options available to achieve a decent throughput:</p>
<ul>
<li>Multiple, bonded (<a href="http://en.wikipedia.org/wiki/Link_aggregation">802.3ad</a>) gigabit links &#8211;  cheap-ish, but some multiport adapters really aren&#8217;t cheap.</li>
<li>4Gbit <a href="http://en.wikipedia.org/wiki/FibreChannel">FibreChannel</a> &#8211; readily available Solaris support, but over-shadowed by 10GigE/Infiniband and requires costly HBAs with <em>extremely</em> expensive XFP/SFP+ modules.</li>
<li><a href="http://en.wikipedia.org/wiki/Infiniband">Infiniband</a> (SDR 4x, 10Gbit) &#8211; really, really cool, but there&#8217;s a huge lack of support in Solaris.
	</li>
<li><a href="http://en.wikipedia.org/wiki/10_Gigabit_Ethernet">10Gigabit Ethernet</a> &#8211; very new, and switches are <strong>extremely</strong> expensive (laughably so, think $20,000 for a 24-port switch + Gbics!) mainly due to the lack of 10GBase-T support (meaning we need 10Base-CX4 or some Fiber-based solution.)</li>
</ul>
<p>So what&#8217;s the answer? We&#8217;re not a Fortune 500 company, so most of this is still out of reach. On top of it all, we need to rely on Solaris for ZFS &#8211; an operating system which seems to have very little manufacturer support, despite its presence in the cluster and virtualisation markets. Sun&#8217;s <a href="http://www.sun.com/bigadmin/hcl">Hardware-Compatibility List</a> is almost devoid of recent Infiniband/10GBase-T adapters, particularly in PCI-E interconnect guises.</p>
<p>It wouldn&#8217;t be so bad if some manufacturer had thought to release a small-scale, 8-10 port 10GigE, 10GBase-T switch. They just don&#8217;t exist.. At present, it&#8217;s quite likely that we&#8217;ll have to dump the idea of a switched fabric altogether, opting instead of multiple point-to-point links.</p>
<p>It seems we&#8217;re either just a few years ahead of ourselves, or really, really out of our depth. </p>
]]></content:encoded>
			<wfw:commentRss>http://www.ninjabadger.net/2008/04/28/crossing-the-gigabit-barrier/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
	</channel>
</rss>

