<?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: The cloud’s most important equation</title>
	<atom:link href="http://www.bitcurrent.com/the-clouds-most-important-equation/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.bitcurrent.com/the-clouds-most-important-equation/</link>
	<description>Just another WordPress site</description>
	<lastBuildDate>Mon, 21 Nov 2011 22:52:22 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.2.1</generator>
	<item>
		<title>By: Why elasticity, performance, and analytics will change how Webops is judged &#124; Bitcurrent</title>
		<link>http://www.bitcurrent.com/the-clouds-most-important-equation/#comment-537</link>
		<dc:creator>Why elasticity, performance, and analytics will change how Webops is judged &#124; Bitcurrent</dc:creator>
		<pubDate>Tue, 26 Oct 2010 21:16:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.bitcurrent.com/?p=309#comment-537</guid>
		<description>[...] elastic computing. Which entails getting billed per capacity used. We&#8217;ve covered this in an earlier post on Bitcurrent, but it seemed the Velocity audience was the perfect place to bring it [...]</description>
		<content:encoded><![CDATA[<p>[...] elastic computing. Which entails getting billed per capacity used. We&#8217;ve covered this in an earlier post on Bitcurrent, but it seemed the Velocity audience was the perfect place to bring it [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: James</title>
		<link>http://www.bitcurrent.com/the-clouds-most-important-equation/#comment-536</link>
		<dc:creator>James</dc:creator>
		<pubDate>Thu, 25 Jun 2009 00:47:46 +0000</pubDate>
		<guid isPermaLink="false">http://www.bitcurrent.com/?p=309#comment-536</guid>
		<description>You might want to clarify that you&#039;re only talking about speed and defining it as user experience within a very narrow context.  I&#039;d have to say that, while response time for a server is certainly important to some user perceptions, it&#039;s but one small piece of the overall user experience.</description>
		<content:encoded><![CDATA[<p>You might want to clarify that you&#8217;re only talking about speed and defining it as user experience within a very narrow context.  I&#8217;d have to say that, while response time for a server is certainly important to some user perceptions, it&#8217;s but one small piece of the overall user experience.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Izzy</title>
		<link>http://www.bitcurrent.com/the-clouds-most-important-equation/#comment-535</link>
		<dc:creator>Izzy</dc:creator>
		<pubDate>Tue, 09 Jun 2009 11:14:04 +0000</pubDate>
		<guid isPermaLink="false">http://www.bitcurrent.com/?p=309#comment-535</guid>
		<description>Alistair, when cloud vendor only provides the basic network infrastructure, CPU and storage, you are still left with options to improve performance (end user experience). For example, cloud doesn&#039;t have to provide application delivery controller (fancy load balancer) as user can deploy its own into the cloud just as easy as any other application they are deploying. At the end it is another piece of software. (Except legacy application delivery controllers that are still hardware based and anchor customers down in the past.)</description>
		<content:encoded><![CDATA[<p>Alistair, when cloud vendor only provides the basic network infrastructure, CPU and storage, you are still left with options to improve performance (end user experience). For example, cloud doesn&#8217;t have to provide application delivery controller (fancy load balancer) as user can deploy its own into the cloud just as easy as any other application they are deploying. At the end it is another piece of software. (Except legacy application delivery controllers that are still hardware based and anchor customers down in the past.)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: random_graph</title>
		<link>http://www.bitcurrent.com/the-clouds-most-important-equation/#comment-534</link>
		<dc:creator>random_graph</dc:creator>
		<pubDate>Wed, 19 Nov 2008 15:37:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.bitcurrent.com/?p=309#comment-534</guid>
		<description>I was very confused to read &quot;experience=traffic/capacity&quot;. It should read either 1) &quot;latency=traffic/capacity&quot; or 2) &quot;user experience=capacity /traffic&quot;. Everyone I know defines high user experience as &quot;good&quot; and low user experience as &quot;bad&quot;.

It would be interesting to learn what these cloud services have in the way of &quot;benevolent degradation&quot;: Would you rather deliver horrible service to a few users or fractionally (imperceptibly?) degrade service to all users? There are always instances of server/router/datacenter loss and how the system behaves when capacity approaches traffic level.</description>
		<content:encoded><![CDATA[<p>I was very confused to read &#8220;experience=traffic/capacity&#8221;. It should read either 1) &#8220;latency=traffic/capacity&#8221; or 2) &#8220;user experience=capacity /traffic&#8221;. Everyone I know defines high user experience as &#8220;good&#8221; and low user experience as &#8220;bad&#8221;.</p>
<p>It would be interesting to learn what these cloud services have in the way of &#8220;benevolent degradation&#8221;: Would you rather deliver horrible service to a few users or fractionally (imperceptibly?) degrade service to all users? There are always instances of server/router/datacenter loss and how the system behaves when capacity approaches traffic level.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Interesting Links for November 12th &#8212; dougmcclure.net</title>
		<link>http://www.bitcurrent.com/the-clouds-most-important-equation/#comment-533</link>
		<dc:creator>Interesting Links for November 12th &#8212; dougmcclure.net</dc:creator>
		<pubDate>Thu, 13 Nov 2008 10:09:56 +0000</pubDate>
		<guid isPermaLink="false">http://www.bitcurrent.com/?p=309#comment-533</guid>
		<description>[...] The cloud&#8217;s most important equation - If you build a traditional data center platform for your application, you worry about three variables: The amount of traffic to your site, your capacity to handle that traffic, and the user experience they get, such as latency. The equation looks like this: [...]</description>
		<content:encoded><![CDATA[<p>[...] The cloud&rsquo;s most important equation &#8211; If you build a traditional data center platform for your application, you worry about three variables: The amount of traffic to your site, your capacity to handle that traffic, and the user experience they get, such as latency. The equation looks like this: [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: tross</title>
		<link>http://www.bitcurrent.com/the-clouds-most-important-equation/#comment-532</link>
		<dc:creator>tross</dc:creator>
		<pubDate>Fri, 24 Oct 2008 20:37:05 +0000</pubDate>
		<guid isPermaLink="false">http://www.bitcurrent.com/?p=309#comment-532</guid>
		<description>Alistaire,
You may remember a company called Ejasent back in those days.  They were the first I remember to use the term elastic computing - certainly predating our use of the term.  Although we didn&#039;t come up with the cool phrase, we were definitely among the first to use service level objectives as a means to control compute resource allocations.

p.s. are you still in touch with Dan Koffler?  He was a great guy to work with back in the early days of Think.</description>
		<content:encoded><![CDATA[<p>Alistaire,<br />
You may remember a company called Ejasent back in those days.  They were the first I remember to use the term elastic computing &#8211; certainly predating our use of the term.  Although we didn&#8217;t come up with the cool phrase, we were definitely among the first to use service level objectives as a means to control compute resource allocations.</p>
<p>p.s. are you still in touch with Dan Koffler?  He was a great guy to work with back in the early days of Think.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

