<?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/"
	xmlns:series="http://unfoldingneurons.com/"
	>

<channel>
	<title>Stephen Foskett, Pack Rat &#187; Alex McDonald Archives  &#8211; Stephen Foskett, Pack Rat</title>
	<atom:link href="http://blog.fosketts.net/tag/alex-mcdonald/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.fosketts.net</link>
	<description>Understanding the accumulation of data</description>
	<lastBuildDate>Fri, 10 Feb 2012 17:40:43 +0000</lastBuildDate>
	<language>en</language>
	<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>Fibre Channel over Token Ring: In-Depth Analysis</title>
		<link>http://blog.fosketts.net/2010/07/16/fibre-channel-token-ring-fcotr/</link>
		<comments>http://blog.fosketts.net/2010/07/16/fibre-channel-token-ring-fcotr/#comments</comments>
		<pubDate>Fri, 16 Jul 2010 16:58:43 +0000</pubDate>
		<dc:creator>Stephen</dc:creator>
				<category><![CDATA[Computer History]]></category>
		<category><![CDATA[Enterprise storage]]></category>
		<category><![CDATA[Personal]]></category>
		<category><![CDATA[Alex McDonald]]></category>
		<category><![CDATA[Cisco]]></category>
		<category><![CDATA[Compellent]]></category>
		<category><![CDATA[Ethan Banks]]></category>
		<category><![CDATA[FCoE]]></category>
		<category><![CDATA[FCoTR]]></category>
		<category><![CDATA[Fibre Channel]]></category>
		<category><![CDATA[Gestalt IT]]></category>
		<category><![CDATA[Greg Ferro]]></category>
		<category><![CDATA[IBM]]></category>
		<category><![CDATA[Jeff Darcy]]></category>
		<category><![CDATA[Liem Nguyen]]></category>
		<category><![CDATA[Madge]]></category>
		<category><![CDATA[Micro Channel]]></category>
		<category><![CDATA[NetApp]]></category>
		<category><![CDATA[Nimble Storage]]></category>
		<category><![CDATA[Packet Pushers]]></category>
		<category><![CDATA[Proteon]]></category>
		<category><![CDATA[Russell Heilling]]></category>
		<category><![CDATA[Steve Duplicity]]></category>
		<category><![CDATA[Tech Field Day]]></category>
		<category><![CDATA[Token Ring]]></category>
		<category><![CDATA[TRILL]]></category>
		<category><![CDATA[W. Curtis Preston]]></category>

		<guid isPermaLink="false">http://blog.fosketts.net/?p=3376</guid>
		<description><![CDATA[Industry veterans fondly remember the LAN wars of the early 1990's, when a diverse set of excellent data link protocols competed for dominance. Although the victory seemed to have gone to Ethernet, industry insiders are looking for a resurgence of better alternatives. One technology, Token Ring, is undeniably superior for transporting modern protocols, especially Fibre Channel for storage. Let's take a look!]]></description>
			<content:encoded><![CDATA[<div id="attachment_3377" class="wp-caption aligncenter" style="width: 310px;  border: 1px solid #dddddd; background-color: #f3f3f3; padding-top: 4px; margin: 10px; text-align:center; display: block; margin-right: auto; margin-left: auto;"><a href="http://blog.fosketts.net/wp-content/uploads/2010/07/IBM_hermaphroditic_connector.jpg" ><img class="size-medium wp-image-3377" title="IBM_hermaphroditic_connector" src="http://blog.fosketts.net/wp-content/uploads/2010/07/IBM_hermaphroditic_connector-300x240.jpg" alt="" width="300" height="240" /></a><p style=' padding: 0 4px 5px; margin: 0;'  class="wp-caption-text">Might Token Ring networks make a comeback or will pigs fly? Industry pundits believe they just might!</p></div>
<p>Industry veterans fondly remember the LAN wars of the early 1990&#8242;s, when a diverse set of excellent data link protocols competed for dominance. Although the victory seemed to have gone to Ethernet, industry insiders are looking for a resurgence of better alternatives. One technology, Token Ring, is undeniably superior for transporting modern protocols, especially Fibre Channel for storage. Let&#8217;s take a look!</p>
<h3>Meeting Modern Needs</h3>
<p>Ethernet is poorly-suited for modern datacenter needs. It lacks the reliability and guaranteed delivery needed by storage protocols, the packets are too small, traffic cannot be prioritized, and thick coaxial cables barrel connectors have proven unreliable. We have heard rumors of so-called &#8220;data center&#8221; enhancements, but analysts are unimpressed. &#8220;Ethernet doesn&#8217;t belong in the data center,&#8221; says Steve Duplicity of the Universal Strategy Federation. &#8220;Data center bridging extensions are as useless as the snow tires my Ferrari dealer sells. No one would ever use them in anger.&#8221;</p>
<p>Token Ring, in contrast, comes prepared for data center use right out of the box. Token Priority, previously used only to avoid collisions, can also be used to prioritize storage traffic over all other payloads. &#8220;Collisions are as common on Ethernet networks as on High School parking lots,&#8221; continues Duplicity, &#8220;and avoidance algorithms work as well as the American drivers&#8217; education system.&#8221; Prioritization of Ethernet packets has been similarly problematic. &#8220;Guaranteed bandwidth is key,&#8221; claims <a href="http://packetpushers.net"  target="_blank">Packet Pusher</a> <a href="http://twitter.com/ecbanks"  target="_blank">Ethan Banks</a>. &#8220;Storage, like school kids, wants to be transported on a bus, and Token Ring can arbitrate that access.&#8221;</p>
<p>Token Ring&#8217;s 20% payload size advantage makes the network more efficient than Ethernet. Although a typical Fibre Channel frame of over 2 KB would still need to be split in two, Token Ring advocates insist that two of their frames will be &#8220;much better&#8221; than two Ethernet frames. This makes Fibre Channel over Token Ring (FCoTR) &#8220;a sure thing&#8221; according to insiders like <a href="http://twitter.com/wcpreston"  target="_blank">W. Curtis Preston</a>.</p>
<h3>Industry Support</h3>
<p>The entire industry is lining up to support Fibre Channel over Token Ring. &#8220;We&#8217;re looking at it,&#8221; admitted <a href="http://twitter.com/liemnguyen"  target="_blank">Liem Nguyen</a> of <a href="http://compellent.com"  target="_blank">Compellent</a> at <a href="http://gestaltit.com/field-day/"  target="_blank">Gestalt IT&#8217;s Tech Field Day</a>. &#8220;The more we hear about FCoTR, the more people are talking about it.&#8221; Nguyen admitted that Compellent was giving FCoTR the consideration it deserved.</p>
<p>Support for the <a href="http://twitter.com/fcotrusergroup"  target="_blank">Fibre Channel over Token Ring user group</a> and blog has jumped rapidly in recent days (see <a rel="nofollow" href="http://fctr.blogspot.com/2010/07/too-good-to-be-true.html" >Too good to be true</a>). Storage startup <a href="http://nimblestorage.com"  target="_blank">Nimble Storage</a> admitted at Tech Field Day that they were watching FCoTR &#8220;with great interest&#8221; and could, in theory, support the protocol alongside iSCSI. <a href="http://netapp.com"  target="_blank">NetApp</a> will respond to &#8220;customer demand&#8221; for FCoTR, agrees <a href="http://twitter.com/alextangent"  target="_blank">Alex McDonald</a>.</p>
<p>Network manufacturers are excited to bring Token Ring back as well. &#8220;Madge Networks is ready to support demand,&#8221; notes USF&#8217;s Duplicity. &#8220;And I hear Proteon is making a comeback as well. IBM, of course, never stopped supporting Token Ring and has thousands of loyal customers.&#8221; Intel is reportedly working on a Micro Channel to PCI Express bridge to ease the transition.</p>
<p>Cisco is predictably sour on the product. &#8220;You have got to be kidding me,&#8221; said a company spokesperson who asked not to be identified. &#8220;Token Ring? Seriously? Are you calling on a Motorola StarTAC?&#8221;</p>
<p>Network experts were cautiously optimistic. &#8220;Trill uses isis, which was largely ripped off from DECnet,&#8221; notes <a href="http://www.twitter.com/xchewtoyx"  target="_blank">potato fan Russell Heilling</a>. &#8220;Is storage over DECnet just around the corner?&#8221; <a href="http://twitter.com/obdurodon"  target="_blank">Jeff Darcy</a> and <a href="http://twitter.com/Etherealmind"  target="_blank">Greg &#8220;Ethereal Mind&#8221; Ferro</a> were excited at this prospect. &#8220;Whoa, the storage folks would LOVE THAT,&#8221; says Ferro of FCoDECnet.</p>
<p>Ferro sums up the appeal of FCoTR. &#8220;Anyone stupid enough to believe in Fibre Channel is absolutely stupid enough to be convinced that FCoTR is a good idea.&#8221;</p>
<p><em>Image source: </em><a rel="nofollow" href="http://en.wikipedia.org/wiki/User:Redgrittybrick" ><em>Redgrittybrick</em></a><em>/Wikimedia</em></p>
<div id="crp_related"><h3>You might also want to read these other posts...</h3><ul><li><a href="http://blog.fosketts.net/2010/07/19/fcotr-exposes-weaknesses-ethernet/"  rel="bookmark" class="crp_title">The FCoTR Phenomenon Exposes the Weaknesses in Ethernet</a></li><li><a href="http://blog.fosketts.net/2011/04/01/vsphere-5-include-vstorage-api-token-ring-integration/"  rel="bookmark" class="crp_title">vSphere 5 To Include vStorage API for Token Ring Integration</a></li><li><a href="http://blog.fosketts.net/2011/01/23/pile-interesting-links-january-21-2011/"  rel="bookmark" class="crp_title">Back From the Pile: Interesting Links, January 21, 2011</a></li><li><a href="http://blog.fosketts.net/2011/10/04/storage-networking-world-snw-orlando/"  rel="bookmark" class="crp_title">Storage Networking World (SNW), Orlando</a></li><li><a href="http://blog.fosketts.net/2011/04/01/pile-interesting-links-april-1-2011/"  rel="bookmark" class="crp_title">Back From the Pile: Interesting Links, April 1, 2011</a></li></ul></div><script src="http://feeds.feedburner.com/~s/sfoskett?i=http://blog.fosketts.net/2010/07/16/fibre-channel-token-ring-fcotr/" type="text/javascript" charset="utf-8"></script><hr />
<p><small>© sfoskett for <a href="http://blog.fosketts.net">Stephen Foskett, Pack Rat</a>, 2010. |
<a href="http://blog.fosketts.net/2010/07/16/fibre-channel-token-ring-fcotr/">Fibre Channel over Token Ring: In-Depth Analysis</a>
<br/>
This post was categorized as <a href="http://blog.fosketts.net/category/everything/computerhistory/" title="View all posts in Computer History" rel="category tag">Computer History</a>, <a href="http://blog.fosketts.net/category/everything/enterprisestorage/" title="View all posts in Enterprise storage" rel="category tag">Enterprise storage</a>, <a href="http://blog.fosketts.net/category/everything/personal/" title="View all posts in Personal" rel="category tag">Personal</a>. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.<br/>
</small></p>]]></content:encoded>
			<wfw:commentRss>http://blog.fosketts.net/2010/07/16/fibre-channel-token-ring-fcotr/feed/</wfw:commentRss>
		<slash:comments>2</slash:comments>
		</item>
		<item>
		<title>The Difference Between &#8220;Integration&#8221; and &#8220;Frankenstein&#8221;</title>
		<link>http://blog.fosketts.net/2009/02/05/difference-integration-frankenstein/</link>
		<comments>http://blog.fosketts.net/2009/02/05/difference-integration-frankenstein/#comments</comments>
		<pubDate>Thu, 05 Feb 2009 15:46:44 +0000</pubDate>
		<dc:creator>Stephen</dc:creator>
				<category><![CDATA[Enterprise storage]]></category>
		<category><![CDATA[Gestalt IT]]></category>
		<category><![CDATA[Alex McDonald]]></category>
		<category><![CDATA[Barry Whyte]]></category>
		<category><![CDATA[Celerra]]></category>
		<category><![CDATA[Chuck Hollis]]></category>
		<category><![CDATA[CLARiiON]]></category>
		<category><![CDATA[DMX]]></category>
		<category><![CDATA[EMC]]></category>
		<category><![CDATA[flash]]></category>
		<category><![CDATA[IBM]]></category>
		<category><![CDATA[integration]]></category>
		<category><![CDATA[NetApp]]></category>
		<category><![CDATA[OEM]]></category>
		<category><![CDATA[PAM]]></category>
		<category><![CDATA[Quantum]]></category>
		<category><![CDATA[RamSan]]></category>
		<category><![CDATA[SSD]]></category>
		<category><![CDATA[STEC]]></category>
		<category><![CDATA[Storagebod]]></category>
		<category><![CDATA[Storagezilla]]></category>
		<category><![CDATA[Texas Memory Systems]]></category>
		<category><![CDATA[Tim Burton]]></category>
		<category><![CDATA[TMS]]></category>
		<category><![CDATA[V-Series]]></category>
		<category><![CDATA[WAFL]]></category>

		<guid isPermaLink="false">http://blog.fosketts.net/?p=1383</guid>
		<description><![CDATA[When is a solution integrated and when is it a Frankenstein-like mashup of tangled tech? Apparently, that line is crossed when it&#8217;s your competitor&#8217;s offering&#8230; In my time in the storage industry, I&#8217;ve seen enough franken-storage come and go to make me skeptical whenever a new &#8220;integrated&#8221; solution is announced. But a lot of this [...]]]></description>
			<content:encoded><![CDATA[<div id="attachment_1384" class="wp-caption alignright" style="width: 151px;  border: 1px solid #dddddd; background-color: #f3f3f3; padding-top: 4px; margin: 10px; text-align:center; float: right;"><a href="http://blog.fosketts.net/wp-content/uploads/2009/02/frankenweenie.jpg" ><img class="size-full wp-image-1384" title="frankenweenie" src="http://blog.fosketts.net/wp-content/uploads/2009/02/frankenweenie.jpg" alt="Frankenweenie saves young Victor in Tim Burton's macabre short film" width="141" height="215" /></a><p style=' padding: 0 4px 5px; margin: 0;'  class="wp-caption-text">Frankenweenie saves young Victor in Tim Burton&#39;s macabre short film</p></div>
<p>When is a solution integrated and when is it a Frankenstein-like mashup of tangled tech? Apparently, that line is crossed <strong>when it&#8217;s your competitor&#8217;s offering</strong>&#8230;</p>
<p>In my time in the storage industry, I&#8217;ve seen enough franken-storage come and go to make me skeptical whenever a new &#8220;integrated&#8221; solution is announced. But a lot of this stuff works just fine, so I also know that <strong>integrated solutions aren&#8217;t always bad</strong>!</p>
<p>The latest industry blog flame war centers around <a href="http://www.netapp.com/us/company/news/news-rel-20090203-flash-ssd.html"  target="_blank">NetApp&#8217;s recently-announced solid state storage solution</a>, which pairs a <a href="http://www.netapp.com/us/products/storage-systems/v3100/"  target="_blank">V-Series NAS head</a> and a Texas Memory Systems <a href="http://www.superssd.com/products/ramsan-500/"  target="_blank">RamSan-500</a> flash storage system. Perhaps NetApp&#8217;s <a href="http://blogs.netapp.com/exposed/" >Val Bercovici</a> did get a bit over-excited in <a href="http://blogs.netapp.com/exposed/2009/02/solid-state-sto.html"  target="_blank">his post on the topic</a>, but he wasn&#8217;t just talking about the RamSan: <strong>He was laying out how NetApp&#8217;s WAFL technology can work in an SSD world</strong>, and using some recent performance test numbers on that solution as well as their PAM cache cards as an illustration of this.</p>
<p>The next thing you know, we have EMC&#8217;s <a rel="nofollow" href="http://storagezilla.typepad.com/storagezilla/2009/02/but-wait-theres-less.html"  target="_blank">Storagezilla</a> and IBM&#8217;s <a href="http://www.ibm.com/developerworks/blogs/page/storagevirtualization?entry=did_it_need_a_press"  target="_blank">Barry Whyte</a> calling the company out for what they (and others. like <a rel="nofollow" href="http://storagebod.typepad.com/storagebods_blog/2009/02/is-that-it.html"  target="_blank">Storagebod</a>) see as an underwhelming product offering. That&#8217;s all well and good, and I&#8217;ll let the reader decide if NetApp&#8217;s moves warranted a press release, but now things have gotten <a href="http://blogs.netapp.com/extensible_netapp/"  target="_blank">uglier</a>&#8230;<span id="more-1383"></span></p>
<p>EMC&#8217;s Chuck Hollis called the whole RamSan idea to account, saying it was &#8220;<a href="http://chucksblog.emc.com/chucks_blog/2009/02/whither-frankenstorage.html"  target="_blank">Frankenstorage</a>&#8220;, causing NetApp&#8217;s Alex MacDonald to engage in a little &#8220;<a href="http://blogs.netapp.com/shadeofblue/2009/02/much-of-the-mai.html"  target="_blank">I know you are but what am I</a>&#8221; in reference to EMC&#8217;s CLARiiON/Celerra &#8220;unified storage&#8221; solutions.</p>
<p>It&#8217;s time to<strong> bring some sanity</strong> to this whole integrated solution concept. Every product in the storage world is an amalgamation of OEM parts to one extent or another, and there are always <a rel="nofollow" href="http://stevetodd.typepad.com/my_weblog/2009/02/xam-from-bleeding-to-cutting-edge.html"  target="_blank">integration issues</a>. Certainly many of EMC&#8217;s offerings could be the subject of name-calling: They use <a rel="nofollow" href="http://thestorageanarchist.typepad.com/weblog/2008/09/1025-flash-wars.html"  target="_blank">STEC SSD drives in the DMX</a>, they use <a rel="nofollow" href="http://thebackupblog.typepad.com/thebackupblog/2008/06/not-just-a-river-in-egypt.html"  target="_blank">Quantum deduplication engines</a> in their CDLs, and their Celerra NS platform <em>does</em> include <a rel="nofollow" href="http://chucksblog.typepad.com/chucks_blog/2008/08/emc-unified-sto.html"  target="_blank">a complete Fibre Channel SAN</a> behind the curtain. But they&#8217;re not alone, and not even wrong in doing this: Every vendor relies on OEMs, and as <a rel="nofollow" href="http://storagezilla.typepad.com/storagezilla/2008/11/mr-backup-gets-it-wrong.html"  target="_blank">a wise man said</a>, &#8220;<strong>working with an OEM gives you the flexibility to pick best of breed technologies</strong>&#8221; and that&#8217;s exactly what customers want. Any objective person would welcome qualification and integration of TMS&#8217; RamSan with a solid platform like the NetApp V-Series &#8211; it&#8217;s a certifiable win for the customer. Just like they would be happy to see EMC leveraging great technology from Quantum and STEC.</p>
<p>Chuck goes on to point out some downsides to these OEM combinations, and they&#8217;re certainly fair criticisms:</p>
<ul>
<li>When you&#8217;re buying this from this guy and that from that guy, <strong>it&#8217;s bound to cost more</strong> because <a rel="nofollow" href="http://storagezilla.typepad.com/storagezilla/2009/01/we-dont-do-free-frank.html"  target="_blank">everyone needs their cut</a>.</li>
<li>Since all attempts at unified heterogeneous device management <a rel="nofollow" href="http://storagearchitect.blogspot.com/2009/01/storage-management-aperi-its-all-over.html"  target="_blank">have failed</a>, a combo is certainly <strong>harder to manage</strong> than a single device.</li>
<li>With multiple vendors in the mix, fingerpointing is common once <strong>support is needed</strong>.</li>
</ul>
<p>But these criticisms can be mitigated by the vendors themselves. They can give up some margin in order to gain market share. They can create unified management interfaces for the combinations they sell and support. And they can really support what they sell, refusing to give in to the temptation to say &#8220;not my problem&#8221; when the going gets rough. <strong>And companies deal with these problems all the time</strong>! Frankenstorage doesn&#8217;t have to be so scary&#8230;</p>
<blockquote><p>This post can also be found on <a href="http://gestaltit.com"  target="_blank">Gestalt IT</a>: <a href="http://gestaltit.com/tech/storage/stephen/the-difference-between-%e2%80%9cintegration%e2%80%9d-and-%e2%80%9cfrankenstein%e2%80%9d/" >The Difference Between “Integration” and “Frankenstein”</a></p>
</blockquote>
<div id="crp_related"><h3>You might also want to read these other posts...</h3><ul><li><a href="http://blog.fosketts.net/2009/04/14/emc-symmetrix-vmax-launch/"  rel="bookmark" class="crp_title">Tracking EMC&#8217;s Symmetrix V-Max Launch</a></li><li><a href="http://blog.fosketts.net/2008/03/12/de-duplication-goes-mainstream/"  rel="bookmark" class="crp_title">De-Duplication Goes Mainstream</a></li><li><a href="http://blog.fosketts.net/2008/11/07/emc-maui/"  rel="bookmark" class="crp_title">EMC About To Take Us To Maui&#8230;</a></li><li><a href="http://blog.fosketts.net/2008/12/12/emulated-fibre-channel-virtualization/"  rel="bookmark" class="crp_title">Of Emulated Fibre Channel, Virtualization, And The Right Tool For The Job</a></li><li><a href="http://blog.fosketts.net/2009/01/26/essential-vmware-esx-iscsi/"  rel="bookmark" class="crp_title">Essential Reading for VMware ESX iSCSI Users!</a></li></ul></div><script src="http://feeds.feedburner.com/~s/sfoskett?i=http://blog.fosketts.net/2009/02/05/difference-integration-frankenstein/" type="text/javascript" charset="utf-8"></script><hr />
<p><small>© sfoskett for <a href="http://blog.fosketts.net">Stephen Foskett, Pack Rat</a>, 2009. |
<a href="http://blog.fosketts.net/2009/02/05/difference-integration-frankenstein/">The Difference Between &#8220;Integration&#8221; and &#8220;Frankenstein&#8221;</a>
<br/>
This post was categorized as <a href="http://blog.fosketts.net/category/everything/enterprisestorage/" title="View all posts in Enterprise storage" rel="category tag">Enterprise storage</a>, <a href="http://blog.fosketts.net/category/gestaltit/" title="View all posts in Gestalt IT" rel="category tag">Gestalt IT</a>. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.<br/>
</small></p>]]></content:encoded>
			<wfw:commentRss>http://blog.fosketts.net/2009/02/05/difference-integration-frankenstein/feed/</wfw:commentRss>
		<slash:comments>2</slash:comments>
		</item>
		<item>
		<title>Of Emulated Fibre Channel, Virtualization, And The Right Tool For The Job</title>
		<link>http://blog.fosketts.net/2008/12/12/emulated-fibre-channel-virtualization/</link>
		<comments>http://blog.fosketts.net/2008/12/12/emulated-fibre-channel-virtualization/#comments</comments>
		<pubDate>Fri, 12 Dec 2008 17:00:44 +0000</pubDate>
		<dc:creator>Stephen</dc:creator>
				<category><![CDATA[Computer History]]></category>
		<category><![CDATA[Enterprise storage]]></category>
		<category><![CDATA[Gestalt IT]]></category>
		<category><![CDATA[Virtual Storage]]></category>
		<category><![CDATA[3PAR]]></category>
		<category><![CDATA[Alex McDonald]]></category>
		<category><![CDATA[Apple]]></category>
		<category><![CDATA[cache]]></category>
		<category><![CDATA[Celerra]]></category>
		<category><![CDATA[Chuck Hollis]]></category>
		<category><![CDATA[CIFS]]></category>
		<category><![CDATA[CLARiiON]]></category>
		<category><![CDATA[Compellent]]></category>
		<category><![CDATA[Data General]]></category>
		<category><![CDATA[Dell]]></category>
		<category><![CDATA[EMC]]></category>
		<category><![CDATA[EqualLogic]]></category>
		<category><![CDATA[FAS]]></category>
		<category><![CDATA[Fibre Channel]]></category>
		<category><![CDATA[HADA]]></category>
		<category><![CDATA[iSCSI]]></category>
		<category><![CDATA[Kostadis Russos]]></category>
		<category><![CDATA[NetApp]]></category>
		<category><![CDATA[NFS]]></category>
		<category><![CDATA[RAID]]></category>
		<category><![CDATA[Storagebod]]></category>
		<category><![CDATA[Symmetrix]]></category>
		<category><![CDATA[Tony Asaro]]></category>
		<category><![CDATA[WAFL]]></category>

		<guid isPermaLink="false">http://blog.fosketts.net/?p=1226</guid>
		<description><![CDATA[EMC&#8217;s Chuck Hollis is one smart guy, and a very verbose blogger. As usual, he sparked a bit of a storm recently when comparing unified storage on EMC&#8217;s Celerra NX4 to NetApp&#8217;s multiprotocol FAS2020 filer. But it was one phrase in particular that got the attention of Alex McDonald and Kostadis Russos of NetApp, Martin/Storagebod, and Tony [...]]]></description>
			<content:encoded><![CDATA[<p>EMC&#8217;s Chuck Hollis is one smart guy, and a very verbose blogger. As usual, he sparked a bit of a storm recently when <a href="http://chucksblog.emc.com/chucks_blog/2008/12/dell-emc-and-the-new-nx4.html"  target="_blank">comparing unified storage on EMC&#8217;s Celerra NX4 to NetApp&#8217;s multiprotocol FAS2020 filer</a>. But it was one phrase in particular that got the attention of <a href="http://blogs.netapp.com/shadeofblue/2008/12/mad-blog-the-e.html"  target="_blank">Alex McDonald</a> and <a href="http://blogs.netapp.com/extensible_netapp/2008/12/mad-blog-chuck.html"  target="_blank">Kostadis Russos</a> of NetApp, <a rel="nofollow" href="http://storagebod.typepad.com/"  target="_blank">Martin/Storagebod</a>, and <a href="http://contemplatingit.com/"  target="_blank">Tony Asaro</a>: &#8220;just because a vendor says they can emulate FC SAN behavior doesn&#8217;t mean it&#8217;s a real FC SAN.&#8221;</p>
<p>What was he getting at? Read <a href="http://chucksblog.emc.com/chucks_blog/2008/12/dell-emc-and-the-new-nx4.html"  target="_blank">the comments</a> in Chuck&#8217;s post and you&#8217;ll understand his implication: Chuck suggests that NetApp &#8220;emulates&#8221; Fibre Channel in their FAS/OnTap devices on top of the WAFL &#8220;file system&#8221;, while EMC&#8217;s AX storage (behind the NX4) uses &#8220;real&#8221; Fibre Channel, so it&#8217;s better. He goes on to say that EMC is doing a brisk business replacing misfit NetApp FC arrays with real FC kit from EMC. But, as is so often the case, the truth is a little more complex than this: <strong>All enterprise storage arrays &#8220;emulate&#8221; Fibre Channel drives to one extent or another, and using the wrong tool for the job will always lead to trouble</strong>.</p>
<p><span id="more-1226"></span></p>
<h3 class="post-subhead">Is It Real Or Is It Virtual?</h3>
<p>Let&#8217;s knock one thing out right away: Chuck is right, <strong>NetApp </strong><em><strong>is</strong></em><strong> emulating Fibre Channel drives</strong> with their FAS arrays. They really <em>do</em> lay out chunks of storage on something kind of like a file system with a bunch of logic mixed in and then pretend that this space is a plain-jane SCSI drive connected over Fibre Channel. And I&#8217;ll add to the &#8220;scandal&#8221; by pointing out that NetApp does exactly the same thing with their iSCSI drives!</p>
<p>Now let&#8217;s move on to an even more important point: <strong>All modern storage arrays emulate SCSI drives</strong>! That&#8217;s right, every enterprise storage array is lying, pretending to serve up basic drives but really slicing and dicing them in the background for their own nefarious purposes!</p>
<p>Who is responsible for this deceit? I place the blame on a few: <a rel="nofollow" href="http://www.eecs.berkeley.edu/Pubs/TechRpts/1987/CSD-87-391.pdf"  target="_blank">Patterson, Gibson, and Katz</a> started the game with their so-called RAID concept, which kicked things off by allowing a few drives to pretend to be a single one. Data General implemented this with cache in their oh-so-clever <a href="http://blog.fosketts.net/2008/11/15/top-ten-storage-hardware/"  target="_blank">HADA</a>, further separating us from The True Disk. But the worst was EMC, with their fully-virtualized Symmetrix array, where there was no definite relationship at all between the LUNs presented to servers and the disks that do all the real work. Some folks would even go so far as to praise this type of <a href="http://blog.fosketts.net/2008/09/14/turning-page-raid/"  target="_self">post-RAID</a> virtualized storage as <a href="http://blog.fosketts.net/2008/11/15/top-ten-storage-hardware/"  target="_blank">innovative</a>!</p>
<p>NetApp takes this &#8220;automated lying&#8221; to the extreme, forcing their innocent hardware to take honest, well-laid-out blocks of <a href="http://blogs.netapp.com/dave/2008/12/is-wafl-a-files.html"  target="_blank">intelligent WAFL space</a> and twist them into vast tracts of dumb pretend-disks. <strong>The nerve!</strong> Compellent, 3PAR, Dell/EqualLogic, and the rest are just as bad, scattering blocks of data willy-nilly across their disks in so-called &#8220;<a href="http://www.storagerap.com/2008/12/basics-of-wide-striping-laying-out-data-in-chunklets.html"  target="_blank">wide stripes</a>&#8220;. But don&#8217;t let Chuck&#8217;s misdirection fool you: EMC is just as guilty with each of their different storage platforms, masquerading as disk drives or file servers and intelligently managing storage underneath! And don&#8217;t get me started on the twisted things VMware does to storage!</p>
<p>Modern? Feh! Let&#8217;s all hope Apple starts producing their <a href="http://blog.fosketts.net/2008/02/19/apple-revs-xsan-and-kills-xserve-raid/"  target="_self">no-feature Xserve RAID</a> again!</p>
<h3 class="post-subhead">Waiting On Angels</h3>
<p>So every modern array emulates disks. What was Chuck&#8217;s point again? Oh yeah, that the AX Fibre Channel storage used by EMC&#8217;s NX4 is superior to the integrated Fibre Channel capability of the NetApp FAS2020! I&#8217;m sure he&#8217;s right for some use cases and wrong for others. FC on the FAS2020 is a perfect match for some, and the NX4/AX wins in a landslide in some circumstances.</p>
<p>The crux of the argument is the fact that NetApp does all sorts of stuff behind the scenes build and support an FC LUN that the EMC AX FC array doesn&#8217;t do. So, although it wouldn&#8217;t be fair to say that one was &#8220;emulated&#8221; and another was not, Chuck would be correct in saying that an FC LUN on an AX is more &#8220;real&#8221; than one on a NetApp FAS. <strong>But arguing over technicalities like this is all <a href="http://www.straightdope.com/columns/read/1008/did-medieval-scholars-argue-over-how-many-angels-could-dance-on-the-head-of-a-pin"  target="_blank">angels and pins</a> and doesn&#8217;t matter in the real world</strong>!</p>
<p>What does matter? In block storage, latency is king. Generally speaking, more cogs and wheels leads to more latency. This is why storage arrays rely so much on large, intelligent caches and vendors are experimenting with all sorts of <a href="http://blogs.netapp.com/storage_nuts_n_bolts/2008/08/performance-acc.html"  target="_blank">cool caching technology</a>. But, ignoring cache, <strong>high-end arrays generally have worse latency than low-end ones</strong> because they have all sorts of translation and virtualization going on in the background. In any I/O situation, increased latency hurts throughput and the perception of performance. And there comes a point when block applications give up waiting and it&#8217;s &#8220;game over, man!&#8221;</p>
<p>I remember migrating from an old CLARiiON 3100 to a brand new Symmetrix 3930 and watching the Symmetrix choke on the incoming data stream. It just couldn&#8217;t write fast enough to handle full streaming reads from the (old-tech) CLARiiON. But once everything was migrated and running, the Symmetrix, with its massive (for the time) 16 GB of cache, widely-spaced data layout, and multiple internal channels, completely destroyed the CLARiiON in real-world performance. This pattern continues today, with devices like the DMX and USP offering much better real-world performance than benchmarks or theoretical techno-arguments would suggest.</p>
<h3 class="post-subhead">So Which Is Better?</h3>
<p>But Chuck and the rest were not talking about high-end stuff here. They are comparing the architecture of entry-level enterprise kit and drawing conclusions about which is best. I personally don&#8217;t care what the internals of the system look like. I care how well it works.</p>
<p>I have personally seen Microsoft Exchange running on low-end FC-connected NetApp FAS arrays, and it worked great. I also helped a customer migrate off of EMC AX that didn&#8217;t give them the performance they needed for their databases. In truth, <strong>lower-end gear is often over-sold</strong> and unable to deliver the performance, features, and reliability specified on data sheets and in vendor presentations.</p>
<p>That&#8217;s right, there&#8217;s more to this picture than raw performance. Consider manageability, for one. NetApp is offering a single-interface integrated system with all protocols (CIFS, NFS, iSCSI, and FC) available from one device. They also offer similar levels of integration for their (really nice) snapshot, replication, and deduplication technology. WAFL is busy doing a lot of great stuff, so I really wouldn&#8217;t be surprised if EMC&#8217;s less-integrated NX/AX offering beats them on performance at the same price point. <strong>Which is more important to you, integration, performance, or features</strong>? And I bet that, if you spent a bit more on a higher-end NetApp box, you could have it all.</p>
<p>On the flip side, EMC is offering a really compelling entry-enterprise combination at a nice price point. The latest NX should be on everyone&#8217;s NAS short list, and I&#8217;m sure the simple FC of the AX array would work well in a smallish Exchange, VMware, or SQL Server environment. It&#8217;s not as unified as NetApp&#8217;s offering management- or feature-wise, but it&#8217;s still pretty good.</p>
<p>Pick the right tool for the job, though. Neither the NX4 nor the FAS2020 is a good fit for a high-I/O application, and that&#8217;s a fact!</p>
<blockquote><p>This post can also be found on <a href="http://gestaltit.com"  target="_blank">Gestalt IT</a>: <a href="http://gestaltit.com/tech/storage/stephen/of-emulated-fibre-channel-virtualization-and-the-right-tool-for-the-job/" >Of Emulated Fibre Channel, Virtualization, And The Right Tool For The Job</a></p>
</blockquote>
<div id="crp_related"><h3>You might also want to read these other posts...</h3><ul><li><a href="http://blog.fosketts.net/2008/08/28/grapples-tangelos-impossible-compare-fairly/"  rel="bookmark" class="crp_title">Grapples and Tangelos: Why it&#8217;s Impossible to Compare Fairly</a></li><li><a href="http://blog.fosketts.net/2009/02/05/difference-integration-frankenstein/"  rel="bookmark" class="crp_title">The Difference Between &#8220;Integration&#8221; and &#8220;Frankenstein&#8221;</a></li><li><a href="http://blog.fosketts.net/2008/07/18/which-storage-protocol-vmware-2/"  rel="bookmark" class="crp_title">Which Storage Protocol For VMware?</a></li><li><a href="http://blog.fosketts.net/2008/10/16/fcoe-versus-iscsi/"  rel="bookmark" class="crp_title">Is the FCoE Starting Pistol Aimed at iSCSI?</a></li><li><a href="http://blog.fosketts.net/2008/02/22/vmware-storage-tidbits/"  rel="bookmark" class="crp_title">VMware Storage Tidbits</a></li></ul></div><script src="http://feeds.feedburner.com/~s/sfoskett?i=http://blog.fosketts.net/2008/12/12/emulated-fibre-channel-virtualization/" type="text/javascript" charset="utf-8"></script><hr />
<p><small>© sfoskett for <a href="http://blog.fosketts.net">Stephen Foskett, Pack Rat</a>, 2008. |
<a href="http://blog.fosketts.net/2008/12/12/emulated-fibre-channel-virtualization/">Of Emulated Fibre Channel, Virtualization, And The Right Tool For The Job</a>
<br/>
This post was categorized as <a href="http://blog.fosketts.net/category/everything/computerhistory/" title="View all posts in Computer History" rel="category tag">Computer History</a>, <a href="http://blog.fosketts.net/category/everything/enterprisestorage/" title="View all posts in Enterprise storage" rel="category tag">Enterprise storage</a>, <a href="http://blog.fosketts.net/category/gestaltit/" title="View all posts in Gestalt IT" rel="category tag">Gestalt IT</a>, <a href="http://blog.fosketts.net/category/everything/virtualstorage/" title="View all posts in Virtual Storage" rel="category tag">Virtual Storage</a>. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.<br/>
</small></p>]]></content:encoded>
			<wfw:commentRss>http://blog.fosketts.net/2008/12/12/emulated-fibre-channel-virtualization/feed/</wfw:commentRss>
		<slash:comments>21</slash:comments>
		</item>
	</channel>
</rss>

