<?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: Compellent CEO needs an Amazon Kindle</title>
	<atom:link href="http://thesantechnologist.com/?feed=rss2&#038;p=146" rel="self" type="application/rss+xml" />
	<link>http://thesantechnologist.com/?p=146</link>
	<description>A Storage Technology Blog by a Technologist</description>
	<lastBuildDate>Mon, 26 Oct 2009 04:20:58 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
	<item>
		<title>By: John</title>
		<link>http://thesantechnologist.com/?p=146&#038;cpage=1#comment-322</link>
		<dc:creator>John</dc:creator>
		<pubDate>Fri, 03 Oct 2008 23:34:26 +0000</pubDate>
		<guid isPermaLink="false">http://thesantechnologist.com/?p=146#comment-322</guid>
		<description>With regards to Compellent, sooner or later customers will get past the illusion of Data Progression and realize that this is the same old legacy controller head and disk shelf architecture that everyone and their mother sells. And what’s worse is that they don’t have a product family like EMC and NetApp do. I don’t know about you, but I wouldn’t want an indeterminate amount of my critical database bits sitting on unreliable SATA disk with no RAID 6 (this is what happens when Data Progression does its thing.) They will say, &quot;just use RAID 10.&quot;; there goes your savings. 

Compellent also has one of the slowest storage systems in the industry because of all the metadata management, software RAID overhead etc, etc, and I challenge them to prove otherwise. They are the kings of marketing, but reality will hit them sooner or later.</description>
		<content:encoded><![CDATA[<p>With regards to Compellent, sooner or later customers will get past the illusion of Data Progression and realize that this is the same old legacy controller head and disk shelf architecture that everyone and their mother sells. And what’s worse is that they don’t have a product family like EMC and NetApp do. I don’t know about you, but I wouldn’t want an indeterminate amount of my critical database bits sitting on unreliable SATA disk with no RAID 6 (this is what happens when Data Progression does its thing.) They will say, &#8220;just use RAID 10.&#8221;; there goes your savings. </p>
<p>Compellent also has one of the slowest storage systems in the industry because of all the metadata management, software RAID overhead etc, etc, and I challenge them to prove otherwise. They are the kings of marketing, but reality will hit them sooner or later.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: John</title>
		<link>http://thesantechnologist.com/?p=146&#038;cpage=1#comment-321</link>
		<dc:creator>John</dc:creator>
		<pubDate>Fri, 03 Oct 2008 23:20:55 +0000</pubDate>
		<guid isPermaLink="false">http://thesantechnologist.com/?p=146#comment-321</guid>
		<description>Chuck left out a few points, like having 1 processor per 12 disks instead of one processor for 100s of disks. Clarion’s throughput bottlenecks are well published. And then there is the rip-and-replace growth strategy instead of just adding nodes. EMC&#039;s usable capacity last I checked:

Software Overhead, Cache Vault &amp; 520 Byte Sectors 5%, RAID Parity (R5 4+1)20%, Spare drives(EMC best practices 1 out of 30)3.3%, Recommended Snap Space Reserve	20%, add remote data Replication (redundancy across sites) with MirrorView/S 50% and usable capacity is 37%.

This is worse than LeftHand configured with multi-site redundancy. The bottom line is that the EMC guys think they know it all, when in fact they are tied to a Data General FLARE &quot;ball-and-chain&quot;, which is good, because we will just continue to eat their lunch, and in a big way going forward.

John</description>
		<content:encoded><![CDATA[<p>Chuck left out a few points, like having 1 processor per 12 disks instead of one processor for 100s of disks. Clarion’s throughput bottlenecks are well published. And then there is the rip-and-replace growth strategy instead of just adding nodes. EMC&#8217;s usable capacity last I checked:</p>
<p>Software Overhead, Cache Vault &amp; 520 Byte Sectors 5%, RAID Parity (R5 4+1)20%, Spare drives(EMC best practices 1 out of 30)3.3%, Recommended Snap Space Reserve	20%, add remote data Replication (redundancy across sites) with MirrorView/S 50% and usable capacity is 37%.</p>
<p>This is worse than LeftHand configured with multi-site redundancy. The bottom line is that the EMC guys think they know it all, when in fact they are tied to a Data General FLARE &#8220;ball-and-chain&#8221;, which is good, because we will just continue to eat their lunch, and in a big way going forward.</p>
<p>John</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Chuck Hollis</title>
		<link>http://thesantechnologist.com/?p=146&#038;cpage=1#comment-318</link>
		<dc:creator>Chuck Hollis</dc:creator>
		<pubDate>Fri, 03 Oct 2008 18:13:30 +0000</pubDate>
		<guid isPermaLink="false">http://thesantechnologist.com/?p=146#comment-318</guid>
		<description>Yep, on target again!!</description>
		<content:encoded><![CDATA[<p>Yep, on target again!!</p>
]]></content:encoded>
	</item>
</channel>
</rss>
