<?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"
	>
<channel>
	<title>Comments on: Google on data center efficiency: Stop making excuses</title>
	<atom:link href="http://itknowledgeexchange.techtarget.com/data-center-facilities/google-on-data-center-efficiency-stop-making-excuses/feed/" rel="self" type="application/rss+xml" />
	<link>http://itknowledgeexchange.techtarget.com/data-center-facilities/google-on-data-center-efficiency-stop-making-excuses/</link>
	<description>A SearchDataCenter.com blog</description>
	<pubDate>Tue, 24 Nov 2009 09:22:33 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.2</generator>
		<item>
		<title>By: Markfontecchio</title>
		<link>http://itknowledgeexchange.techtarget.com/data-center-facilities/google-on-data-center-efficiency-stop-making-excuses/#comment-39</link>
		<dc:creator>Markfontecchio</dc:creator>
		<pubDate>Fri, 17 Apr 2009 20:22:57 +0000</pubDate>
		<guid isPermaLink="false">http://itknowledgeexchange.techtarget.com/data-center-facilities/google-on-data-center-efficiency-stop-making-excuses/#comment-39</guid>
		<description>Jack,

I agree that a measure of productivity per watt is a better metric. You could have a data center with a PUE of 1.2, but if the IT equipment isn't doing anything, it's all a waste of energy anyway. Once an industry-agreed productivity metric comes out, I would guess that data centers with servers running at higher utilization will benefit greatly, even if those servers might consume more power than other servers that are less power hungry overall but run at 10%. We'll see.</description>
		<content:encoded><![CDATA[<p>Jack,</p>
<p>I agree that a measure of productivity per watt is a better metric. You could have a data center with a PUE of 1.2, but if the IT equipment isn&#8217;t doing anything, it&#8217;s all a waste of energy anyway. Once an industry-agreed productivity metric comes out, I would guess that data centers with servers running at higher utilization will benefit greatly, even if those servers might consume more power than other servers that are less power hungry overall but run at 10%. We&#8217;ll see.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jpouchet</title>
		<link>http://itknowledgeexchange.techtarget.com/data-center-facilities/google-on-data-center-efficiency-stop-making-excuses/#comment-38</link>
		<dc:creator>Jpouchet</dc:creator>
		<pubDate>Fri, 17 Apr 2009 20:03:35 +0000</pubDate>
		<guid isPermaLink="false">http://itknowledgeexchange.techtarget.com/data-center-facilities/google-on-data-center-efficiency-stop-making-excuses/#comment-38</guid>
		<description>This is an interesting discussion on comparing different types of data centers, but with its focus on PUE it misses the larger point. PUE is a nice-to-know bit of data, but it’s simply a ratio of power flow within a facility. PUE does not relate to PRODUCTIVITY, which is the single most important business measure for a data center. To understand productivity, we need to measure useful work per watt and perhaps assign values to different types of work performed in the data center. All bits, bytes and applications may not have the same relative value. 

The industry as a whole recognizes this problem and The Green Grid is working to develop a suitable tool to resolve this issue. Until then, there are several proxies for useful work, including Emerson Network Power’s CUPS (Compute Units Per Second) that enable us to have a more meaningful discussion on data center efficiency and overall productivity. http://www.emerson.com/edc/page/Efficiency-Metrics.aspx  . 

Jack Pouchet
Emerson Network Power
http://www.efficientdatacenters.com</description>
		<content:encoded><![CDATA[<p>This is an interesting discussion on comparing different types of data centers, but with its focus on PUE it misses the larger point. PUE is a nice-to-know bit of data, but it’s simply a ratio of power flow within a facility. PUE does not relate to PRODUCTIVITY, which is the single most important business measure for a data center. To understand productivity, we need to measure useful work per watt and perhaps assign values to different types of work performed in the data center. All bits, bytes and applications may not have the same relative value. </p>
<p>The industry as a whole recognizes this problem and The Green Grid is working to develop a suitable tool to resolve this issue. Until then, there are several proxies for useful work, including Emerson Network Power’s CUPS (Compute Units Per Second) that enable us to have a more meaningful discussion on data center efficiency and overall productivity.&nbsp;&lt;a href="http://www.emerson.com/edc/page/Efficiency-Metrics.aspx" title="http://www.emerson.com/edc/page/Efficiency-Metrics.aspx" target="_blank"&gt;http://www.emerson.com/edc/page/Efficien&#8230;&lt;/a&gt;  . </p>
<p>Jack Pouchet<br />
Emerson Network Power<br />
&nbsp;&lt;a href="http://www.efficientdatacenters.com" title="http://www.efficientdatacenters.com" target="_blank"&gt;http://www.efficientdatacenters.com&lt;/a&gt;</p>
]]></content:encoded>
	</item>
</channel>
</rss>
<!-- dynamic -->