<?xml version="1.0" encoding="utf-8"?>
<?xml-stylesheet type="text/css" href="http://commons.oreilly.com/wiki/skins/common/feed.css?97"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
	<channel>
		<title>Talk:Cloud computing perspectives and questions - Revision history</title>
		<link>http://commons.oreilly.com/wiki/index.php?title=Talk:Cloud_computing_perspectives_and_questions&amp;action=history</link>
		<description>Revision history for this page on the wiki</description>
		<language>en</language>
		<generator>MediaWiki 1.11.0</generator>
		<lastBuildDate>Sun, 19 May 2013 22:30:54 GMT</lastBuildDate>
		<item>
			<title>Davecb: New page: Re: Universal Compute Unit (UcU)   This is probably a useful unit, but it should be some form of &quot;balanced&quot; composite, like the old IBM &quot;a mip. a meg, a gig&quot;, meaning 1 MIP of CPU, 1 MB of...</title>
			<link>http://commons.oreilly.com/wiki/index.php?title=Talk:Cloud_computing_perspectives_and_questions&amp;diff=24820&amp;oldid=prev</link>
			<description>&lt;p&gt;New page: Re: Universal Compute Unit (UcU)   This is probably a useful unit, but it should be some form of &amp;quot;balanced&amp;quot; composite, like the old IBM &amp;quot;a mip. a meg, a gig&amp;quot;, meaning 1 MIP of CPU, 1 MB of...&lt;/p&gt;
&lt;p&gt;&lt;b&gt;New page&lt;/b&gt;&lt;/p&gt;&lt;div&gt;Re: Universal Compute Unit (UcU) &lt;br /&gt;
&lt;br /&gt;
This is probably a useful unit, but it should be some form&lt;br /&gt;
of &amp;quot;balanced&amp;quot; composite, like the old IBM &amp;quot;a mip. a meg, a gig&amp;quot;,&lt;br /&gt;
meaning 1 MIP of CPU, 1 MB of memory and 1 GB of disk.&lt;br /&gt;
I suspect this is what you were thinking of...&lt;br /&gt;
&lt;br /&gt;
If it were just a CPU unit, I could easily have a 10 UcU&lt;br /&gt;
system which gave me half the performance of a 5 UcU system&lt;br /&gt;
because the first one was low on I/O bandwidth and the&lt;br /&gt;
second wasn't.&lt;br /&gt;
&lt;br /&gt;
It doesn't have to be a perfect balance: estimating quanta&lt;br /&gt;
based on TPC benchmarks works surprisingly well, even though&lt;br /&gt;
everyone games the benchmarks.  Perhaps '''because''' they all&lt;br /&gt;
game the benchmarks.&lt;br /&gt;
&lt;br /&gt;
--dave&lt;/div&gt;</description>
			<pubDate>Mon, 13 Jul 2009 03:12:10 GMT</pubDate>			<dc:creator>Davecb</dc:creator>			<comments>http://commons.oreilly.com/wiki/index.php/Talk:Cloud_computing_perspectives_and_questions</comments>		</item>
	</channel>
</rss>