<?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: What to do with MySQL Full Text Search while migrating to Innodb ?</title>
	<atom:link href="http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/</link>
	<description>Percona&#039;s MySQL &#38; InnoDB performance and scalability blog</description>
	<lastBuildDate>Sat, 11 Feb 2012 16:45:54 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.2.1</generator>
	<item>
		<title>By: Baron Schwartz</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-711240</link>
		<dc:creator>Baron Schwartz</dc:creator>
		<pubDate>Sun, 17 Jan 2010 22:56:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-711240</guid>
		<description>Unless I&#039;m missing your point, that would solve absolutely nothing.</description>
		<content:encoded><![CDATA[<p>Unless I&#8217;m missing your point, that would solve absolutely nothing.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: heradas</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-711150</link>
		<dc:creator>heradas</dc:creator>
		<pubDate>Sun, 17 Jan 2010 19:00:07 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-711150</guid>
		<description>how about just using mysql cluster? that should the prblem</description>
		<content:encoded><![CDATA[<p>how about just using mysql cluster? that should the prblem</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: peter</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-652545</link>
		<dc:creator>peter</dc:creator>
		<pubDate>Mon, 14 Sep 2009 15:18:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-652545</guid>
		<description>Chritopher,

Yes. Everything has a cost. I would see it same as running memcache or other caching system with MySQL - you increase complexity you get performance.

About Migrations - Sphinx is a separate server and can be moved separately from MySQL.
Also because it pulls data from MySQL you can often move MySQL, when get the new sphinx installation running off that MySQL and when stop using old one.</description>
		<content:encoded><![CDATA[<p>Chritopher,</p>
<p>Yes. Everything has a cost. I would see it same as running memcache or other caching system with MySQL &#8211; you increase complexity you get performance.</p>
<p>About Migrations &#8211; Sphinx is a separate server and can be moved separately from MySQL.<br />
Also because it pulls data from MySQL you can often move MySQL, when get the new sphinx installation running off that MySQL and when stop using old one.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: peter</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-652540</link>
		<dc:creator>peter</dc:creator>
		<pubDate>Mon, 14 Sep 2009 15:13:54 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-652540</guid>
		<description>Jan,

Do not forget MyISAM is pretty much always used for on disk temporary tables if you have Innodb so you never get off balancing problem completely.

Also you need to balance memory between innodb buffer pool and OS cache which is needed for some things, query cache etc.

Also setting it 50-50 is not a good method.  The size and usage is what should drive the value.  If there is single full text search table remains I would like to keep key_buffer same as .MYI takes (or expected to take) for the table.</description>
		<content:encoded><![CDATA[<p>Jan,</p>
<p>Do not forget MyISAM is pretty much always used for on disk temporary tables if you have Innodb so you never get off balancing problem completely.</p>
<p>Also you need to balance memory between innodb buffer pool and OS cache which is needed for some things, query cache etc.</p>
<p>Also setting it 50-50 is not a good method.  The size and usage is what should drive the value.  If there is single full text search table remains I would like to keep key_buffer same as .MYI takes (or expected to take) for the table.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Christopher Grello</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-652536</link>
		<dc:creator>Christopher Grello</dc:creator>
		<pubDate>Mon, 14 Sep 2009 15:05:39 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-652536</guid>
		<description>Since you&#039;re adding another item to the database environment you increase the complexity for day-to-day things, but also migrations. The benefits gained are certainly nice, but I&#039;m curious if anyone has any good practice recommendations for migration Sphinx from one box to another when you need to move MySQL.</description>
		<content:encoded><![CDATA[<p>Since you&#8217;re adding another item to the database environment you increase the complexity for day-to-day things, but also migrations. The benefits gained are certainly nice, but I&#8217;m curious if anyone has any good practice recommendations for migration Sphinx from one box to another when you need to move MySQL.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jan</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-652449</link>
		<dc:creator>Jan</dc:creator>
		<pubDate>Mon, 14 Sep 2009 12:08:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-652449</guid>
		<description>Peter, I assume your suggestion would be to e.g. give each database engine half the ram. Of course, that would work perfectly, but you are wasting precious memory. Having InnoDB and MyISAM &quot;share&quot; memory does not work very well.

So yes, you can do it, but I think other methods are better.</description>
		<content:encoded><![CDATA[<p>Peter, I assume your suggestion would be to e.g. give each database engine half the ram. Of course, that would work perfectly, but you are wasting precious memory. Having InnoDB and MyISAM &#8220;share&#8221; memory does not work very well.</p>
<p>So yes, you can do it, but I think other methods are better.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: peter</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-651710</link>
		<dc:creator>peter</dc:creator>
		<pubDate>Fri, 11 Sep 2009 16:19:16 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-651710</guid>
		<description>Jan,

Mixing MyISAM and Innodb on the same server add balancing challenge but I think it is far from &quot;not a good idea&quot;.  You just need to consider your table sizes and usage pattern to balance between Innodb Buffer Pool size, Key Cache and OS Cache.</description>
		<content:encoded><![CDATA[<p>Jan,</p>
<p>Mixing MyISAM and Innodb on the same server add balancing challenge but I think it is far from &#8220;not a good idea&#8221;.  You just need to consider your table sizes and usage pattern to balance between Innodb Buffer Pool size, Key Cache and OS Cache.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: peter</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-651708</link>
		<dc:creator>peter</dc:creator>
		<pubDate>Fri, 11 Sep 2009 16:17:12 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-651708</guid>
		<description>Imran,

Indeed if you create Shadow MyISAM table you would rarely keep it same as original table.  You probably will not need a lot of indexes as well as many columns.</description>
		<content:encoded><![CDATA[<p>Imran,</p>
<p>Indeed if you create Shadow MyISAM table you would rarely keep it same as original table.  You probably will not need a lot of indexes as well as many columns.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jan</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-651631</link>
		<dc:creator>Jan</dc:creator>
		<pubDate>Fri, 11 Sep 2009 08:55:38 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-651631</guid>
		<description>Mixing InnoDB and MyISAM on a single server is not a good idea, due to memory allocation issues. So the only realistic option is a MyISAM slave. Or go Lucene/Solr.</description>
		<content:encoded><![CDATA[<p>Mixing InnoDB and MyISAM on a single server is not a good idea, due to memory allocation issues. So the only realistic option is a MyISAM slave. Or go Lucene/Solr.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: imran</title>
		<link>http://www.mysqlperformanceblog.com/2009/09/10/what-to-do-with-mysql-full-text-search-while-migrating-to-innodb/comment-page-1/#comment-651612</link>
		<dc:creator>imran</dc:creator>
		<pubDate>Fri, 11 Sep 2009 07:57:40 +0000</pubDate>
		<guid isPermaLink="false">http://www.mysqlperformanceblog.com/?p=1093#comment-651612</guid>
		<description>To add to the above though, I still prefer to use Sphinx for any InnoDB table warranting FT indexing.</description>
		<content:encoded><![CDATA[<p>To add to the above though, I still prefer to use Sphinx for any InnoDB table warranting FT indexing.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

