<?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 for Hack MySQL Blog</title>
	<atom:link href="http://hackmysql.com/blog/comments/feed/" rel="self" type="application/rss+xml" />
	<link>http://hackmysql.com/blog</link>
	<description></description>
	<lastBuildDate>Tue, 30 Aug 2011 14:21:11 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>Comment on New algorithm for calculating 95 percentile by Daniel Nichter</title>
		<link>http://hackmysql.com/blog/2011/08/29/new-algorithm-for-calculating-95-percentile/comment-page-1/#comment-1012</link>
		<dc:creator>Daniel Nichter</dc:creator>
		<pubDate>Tue, 30 Aug 2011 14:21:11 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=222#comment-1012</guid>
		<description>Justin, I think 21k-1-spike5-1 is off because it&#039;s a perfect edge case.  As the filename kind of suggests, the values are:

10k @ 1s
1k @ 5s
10k @ 1s

So the 5s spike values are exactly 5% of total values (1k/20k).  That make the 95%/5% edge something like 1s/5s.  But the 5s values aren&#039;t all 5s, some are 2, 3, or 4.  So the edge is more like 1s/2s.  Then (I should have probably made this explicit in the post), 95% here is actually &quot;95% of values less than &lt;em&gt;or equal to&lt;/em&gt; X&quot; whereas technically a percentile should be just &quot;less than X&quot;.  Using &lt;= is inherited from mk-query-digest, but perhaps I should re-examine if that&#039;s correct.</description>
		<content:encoded><![CDATA[<p>Justin, I think 21k-1-spike5-1 is off because it&#8217;s a perfect edge case.  As the filename kind of suggests, the values are:</p>
<p>10k @ 1s<br />
1k @ 5s<br />
10k @ 1s</p>
<p>So the 5s spike values are exactly 5% of total values (1k/20k).  That make the 95%/5% edge something like 1s/5s.  But the 5s values aren&#8217;t all 5s, some are 2, 3, or 4.  So the edge is more like 1s/2s.  Then (I should have probably made this explicit in the post), 95% here is actually &#8220;95% of values less than <em>or equal to</em> X&#8221; whereas technically a percentile should be just &#8220;less than X&#8221;.  Using &lt;= is inherited from mk-query-digest, but perhaps I should re-examine if that&#039;s correct.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on New algorithm for calculating 95 percentile by Daniel Nichter</title>
		<link>http://hackmysql.com/blog/2011/08/29/new-algorithm-for-calculating-95-percentile/comment-page-1/#comment-1011</link>
		<dc:creator>Daniel Nichter</dc:creator>
		<pubDate>Tue, 30 Aug 2011 14:06:32 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=222#comment-1011</guid>
		<description>Thanks for taking a look Roland.  :-)</description>
		<content:encoded><![CDATA[<p>Thanks for taking a look Roland.  :-)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on New algorithm for calculating 95 percentile by Justin Rovang</title>
		<link>http://hackmysql.com/blog/2011/08/29/new-algorithm-for-calculating-95-percentile/comment-page-1/#comment-1010</link>
		<dc:creator>Justin Rovang</dc:creator>
		<pubDate>Tue, 30 Aug 2011 13:46:02 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=222#comment-1010</guid>
		<description>Is there a reason or conjecture to why the 21k-1-spike5-1 test was so far off?</description>
		<content:encoded><![CDATA[<p>Is there a reason or conjecture to why the 21k-1-spike5-1 test was so far off?</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on New algorithm for calculating 95 percentile by Roland Bouman</title>
		<link>http://hackmysql.com/blog/2011/08/29/new-algorithm-for-calculating-95-percentile/comment-page-1/#comment-1009</link>
		<dc:creator>Roland Bouman</dc:creator>
		<pubDate>Tue, 30 Aug 2011 12:37:33 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=222#comment-1009</guid>
		<description>Hi Daniel! 

wow - this is great, very interesting stuff :)

I&#039;m now downloading your code to check out the algorithm - I&#039;m really curious how it works. 

Oh, to get a quick grasp of the result, I calculated the correlation between (REAL_95,OLD_95) and (REAL_95,NEW_95), and the results are 0,9998926355 and 0,9999572588 respectively - I think it&#039;s safe to say both are accurate enough, and it seels the new one is even a bit more accurate than the old one (for this dataset). 

Exciting :)</description>
		<content:encoded><![CDATA[<p>Hi Daniel! </p>
<p>wow &#8211; this is great, very interesting stuff :)</p>
<p>I&#8217;m now downloading your code to check out the algorithm &#8211; I&#8217;m really curious how it works. </p>
<p>Oh, to get a quick grasp of the result, I calculated the correlation between (REAL_95,OLD_95) and (REAL_95,NEW_95), and the results are 0,9998926355 and 0,9999572588 respectively &#8211; I think it&#8217;s safe to say both are accurate enough, and it seels the new one is even a bit more accurate than the old one (for this dataset). </p>
<p>Exciting :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on MySQL tools article by Daniel Nichter</title>
		<link>http://hackmysql.com/blog/2011/08/01/mysql-tools-article/comment-page-1/#comment-969</link>
		<dc:creator>Daniel Nichter</dc:creator>
		<pubDate>Wed, 10 Aug 2011 15:56:59 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=217#comment-969</guid>
		<description>I&#039;m not aware of any tools for studying MySQL source code, but there is http://forge.mysql.com/wiki/MySQL_Internals.</description>
		<content:encoded><![CDATA[<p>I&#8217;m not aware of any tools for studying MySQL source code, but there is <a href="http://forge.mysql.com/wiki/MySQL_Internals" rel="nofollow">http://forge.mysql.com/wiki/MySQL_Internals</a>.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on MySQL tools article by whitepoplar</title>
		<link>http://hackmysql.com/blog/2011/08/01/mysql-tools-article/comment-page-1/#comment-964</link>
		<dc:creator>whitepoplar</dc:creator>
		<pubDate>Wed, 10 Aug 2011 01:29:21 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=217#comment-964</guid>
		<description>hello

would you like to tell us that is there any good tools for studying MySQL source codes.

As i know ,gdb..and vim.</description>
		<content:encoded><![CDATA[<p>hello</p>
<p>would you like to tell us that is there any good tools for studying MySQL source codes.</p>
<p>As i know ,gdb..and vim.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on MySQL tools article by Shlomi Noach</title>
		<link>http://hackmysql.com/blog/2011/08/01/mysql-tools-article/comment-page-1/#comment-936</link>
		<dc:creator>Shlomi Noach</dc:creator>
		<pubDate>Sun, 07 Aug 2011 05:27:35 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=217#comment-936</guid>
		<description>Daniel,
Thank you! I&#039;m not good with publicizing, but apparently you&#039;re doing the work for me :)</description>
		<content:encoded><![CDATA[<p>Daniel,<br />
Thank you! I&#8217;m not good with publicizing, but apparently you&#8217;re doing the work for me :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on MySQL tools article by Daniel Nichter</title>
		<link>http://hackmysql.com/blog/2011/08/01/mysql-tools-article/comment-page-1/#comment-875</link>
		<dc:creator>Daniel Nichter</dc:creator>
		<pubDate>Tue, 02 Aug 2011 16:10:35 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=217#comment-875</guid>
		<description>Shlomi has a good thing going with oak-security-audit.  He should publicize it more.</description>
		<content:encoded><![CDATA[<p>Shlomi has a good thing going with oak-security-audit.  He should publicize it more.</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on MySQL tools article by George</title>
		<link>http://hackmysql.com/blog/2011/08/01/mysql-tools-article/comment-page-1/#comment-874</link>
		<dc:creator>George</dc:creator>
		<pubDate>Tue, 02 Aug 2011 16:02:54 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=217#comment-874</guid>
		<description>Thanks Daniel, openark kit security audit is a new one for me thanks for the introduction. Added it to my tools I use including my fav mysqlreport :)</description>
		<content:encoded><![CDATA[<p>Thanks Daniel, openark kit security audit is a new one for me thanks for the introduction. Added it to my tools I use including my fav mysqlreport :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>Comment on MySQL tools article by Daniel Nichter</title>
		<link>http://hackmysql.com/blog/2011/08/01/mysql-tools-article/comment-page-1/#comment-871</link>
		<dc:creator>Daniel Nichter</dc:creator>
		<pubDate>Tue, 02 Aug 2011 14:45:05 +0000</pubDate>
		<guid isPermaLink="false">http://hackmysql.com/blog/?p=217#comment-871</guid>
		<description>Thanks, Daniel.  MySQL Sandbox was in my original list of possibilities, but I had to narrow it down to 10 tools and I felt MySQL Sandbox was not a tool strictly speaking because a tool does some job on/to/for MySQL but MySQL Sandbox _is_ MySQL.</description>
		<content:encoded><![CDATA[<p>Thanks, Daniel.  MySQL Sandbox was in my original list of possibilities, but I had to narrow it down to 10 tools and I felt MySQL Sandbox was not a tool strictly speaking because a tool does some job on/to/for MySQL but MySQL Sandbox _is_ MySQL.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

