<?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: System size and Performance Optimization</title>
	<atom:link href="http://www.mysqlperformanceblog.com/2007/02/27/system-size-and-performance-optimization/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.mysqlperformanceblog.com/2007/02/27/system-size-and-performance-optimization/</link>
	<description>Everything about MySQL Performance</description>
	<pubDate>Wed, 01 Oct 2008 00:22:15 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: peter</title>
		<link>http://www.mysqlperformanceblog.com/2007/02/27/system-size-and-performance-optimization/#comment-65400</link>
		<dc:creator>peter</dc:creator>
		<pubDate>Thu, 01 Mar 2007 10:53:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2007/02/27/system-size-and-performance-optimization/#comment-65400</guid>
		<description>Gunnar, 

I agree with you completely. In fact I was proposing for Virtual Columns to be implemented years ago, I guess that Worklog Entry is still located somewhere.  It is very neat feature for practical use but it is not ANSI SQL stuff and MySQL has implementing Standard features as priority now. 

I'm also not sure how easy it would be to implement as it would need to be supported in all storage engines, because you basically index something when you do not store the data. 

Well storage engines of course could have a flag to say they do not support indexes on virtual columns</description>
		<content:encoded><![CDATA[<p>Gunnar, </p>
<p>I agree with you completely. In fact I was proposing for Virtual Columns to be implemented years ago, I guess that Worklog Entry is still located somewhere.  It is very neat feature for practical use but it is not ANSI SQL stuff and MySQL has implementing Standard features as priority now. </p>
<p>I&#8217;m also not sure how easy it would be to implement as it would need to be supported in all storage engines, because you basically index something when you do not store the data. </p>
<p>Well storage engines of course could have a flag to say they do not support indexes on virtual columns</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Gunnar</title>
		<link>http://www.mysqlperformanceblog.com/2007/02/27/system-size-and-performance-optimization/#comment-65395</link>
		<dc:creator>Gunnar</dc:creator>
		<pubDate>Thu, 01 Mar 2007 10:43:59 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/2007/02/27/system-size-and-performance-optimization/#comment-65395</guid>
		<description>Hi,

IMHO, I very simply way to get a bit of performance is reducing size. Try to store the data as compact as possible.

What I miss in MySQL for this is the possibility to use "indexes on virtual columns" or "indexes on functions".
For example often I use an "index on soundex(columns) and index on soundex(reverse(columsn))
Today I have to hold create redundant columns in MySQL to hold soundex of columns to able to create the index.
I think for many projects I could easely save 25% disk and memory size if I could create the indexes without creating the column.
Oher RDMS like Postgres support this.

What is your opinion on this?
Would you think that support for this will be difficult to add to MySQL?

Cheers
Gunnar</description>
		<content:encoded><![CDATA[<p>Hi,</p>
<p>IMHO, I very simply way to get a bit of performance is reducing size. Try to store the data as compact as possible.</p>
<p>What I miss in MySQL for this is the possibility to use &#8220;indexes on virtual columns&#8221; or &#8220;indexes on functions&#8221;.<br />
For example often I use an &#8220;index on soundex(columns) and index on soundex(reverse(columsn))<br />
Today I have to hold create redundant columns in MySQL to hold soundex of columns to able to create the index.<br />
I think for many projects I could easely save 25% disk and memory size if I could create the indexes without creating the column.<br />
Oher RDMS like Postgres support this.</p>
<p>What is your opinion on this?<br />
Would you think that support for this will be difficult to add to MySQL?</p>
<p>Cheers<br />
Gunnar</p>
]]></content:encoded>
	</item>
</channel>
</rss>
