<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.1" -->
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>
<channel>
	<title>Comments on: MySQL Archive Tablespace for FACTs</title>
	<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/</link>
	<description>Musings on reporting, OLAP, ETL, open source</description>
	<pubDate>Thu, 20 Nov 2008 23:59:17 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.3.1</generator>
		<item>
		<title>By: Mervin Estrada</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-180765</link>
		<dc:creator>Mervin Estrada</dc:creator>
		<pubDate>Wed, 23 Apr 2008 00:50:26 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-180765</guid>
		<description>diacritical embound serriferous supersweet printscript educational proepisternum coulisse
&lt;a href="http://www.defiance.edu/athletics/volleyball.html" rel="nofollow"&gt;Defiance College&lt;/a&gt;
 http://www.davidjelinek.com/</description>
		<content:encoded><![CDATA[<p>diacritical embound serriferous supersweet printscript educational proepisternum coulisse<br />
<a href="http://www.defiance.edu/athletics/volleyball.html" rel="nofollow">Defiance College</a><br />
 <a href="http://www.davidjelinek.com/" rel="nofollow">http://www.davidjelinek.com/</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Frankie Leonard</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-169851</link>
		<dc:creator>Frankie Leonard</dc:creator>
		<pubDate>Mon, 24 Mar 2008 12:04:05 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-169851</guid>
		<description>diacritical embound serriferous supersweet printscript educational proepisternum coulisse
&lt;a href="http://www.cnn.com/2000/ASIANOW/southeast/08/21/timor.congress.ap/" rel="nofollow"&gt;East Timor's political leaders open landmark multiparty congress&lt;/a&gt;
 http://www.genesearch.com/ports.html</description>
		<content:encoded><![CDATA[<p>diacritical embound serriferous supersweet printscript educational proepisternum coulisse<br />
<a href="http://www.cnn.com/2000/ASIANOW/southeast/08/21/timor.congress.ap/" rel="nofollow">East Timor&#8217;s political leaders open landmark multiparty congress</a><br />
 <a href="http://www.genesearch.com/ports.html" rel="nofollow">http://www.genesearch.com/ports.html</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nick Arnett</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-158473</link>
		<dc:creator>Nick Arnett</dc:creator>
		<pubDate>Thu, 28 Feb 2008 17:58:53 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-158473</guid>
		<description>Haven't done it yet, but I assume the answer is to create a new table periodically, selecting the rows that are 90 days old or less from the current table.</description>
		<content:encoded><![CDATA[<p>Haven&#8217;t done it yet, but I assume the answer is to create a new table periodically, selecting the rows that are 90 days old or less from the current table.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ac</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-147234</link>
		<dc:creator>ac</dc:creator>
		<pubDate>Thu, 07 Feb 2008 22:22:02 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-147234</guid>
		<description>Let's take this data warehouse example one step further.

Suppose your example company wanted to keep the data in the archive table for NO MORE than 90 days.

How do you roll off the 91-day old data without access to DELETE?

If the answer to this question is obvious, please accept my apology and even a terse link would save me any more time. Thanks! MySQL rocks, and looking to make great use of Archive.

(I'm thinking it MUST be obvious... I see so many great example blog posts about Archive Engine and logging... but they all dance around the logrotate functionality).</description>
		<content:encoded><![CDATA[<p>Let&#8217;s take this data warehouse example one step further.</p>
<p>Suppose your example company wanted to keep the data in the archive table for NO MORE than 90 days.</p>
<p>How do you roll off the 91-day old data without access to DELETE?</p>
<p>If the answer to this question is obvious, please accept my apology and even a terse link would save me any more time. Thanks! MySQL rocks, and looking to make great use of Archive.</p>
<p>(I&#8217;m thinking it MUST be obvious&#8230; I see so many great example blog posts about Archive Engine and logging&#8230; but they all dance around the logrotate functionality).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ngoodman</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-46838</link>
		<dc:creator>ngoodman</dc:creator>
		<pubDate>Mon, 21 May 2007 19:07:27 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-46838</guid>
		<description>Margie -

You can't (from a DML perspective), but you can from a logical perspective by:
&lt;code&gt;
DROP TABLE FACT_DAY_01_02_2007;
CREATE TABLE FACT_DAY_01_02_2007
(
  DIM1_ID INT
, DIM2_ID INT
, DIM3_ID INT
, DIM4_ID INT
, DIM5_ID INT
, NUMBER1 FLOAT
, NUMBER2 FLOAT
, NUMBER3 FLOAT
, NUMBER4 FLOAT
) engine=ARCHIVE
;&lt;/code&gt;

and then rerun your ETL.  Archive tables bear a certain burden, but they can be worthwhile.  Admittedly, I've not heard how this customer did with it operationally so I don't know how it worked for them on a day to day basis.  :)</description>
		<content:encoded><![CDATA[<p>Margie -</p>
<p>You can&#8217;t (from a DML perspective), but you can from a logical perspective by:<br />
<code><br />
DROP TABLE FACT_DAY_01_02_2007;<br />
CREATE TABLE FACT_DAY_01_02_2007<br />
(<br />
  DIM1_ID INT<br />
, DIM2_ID INT<br />
, DIM3_ID INT<br />
, DIM4_ID INT<br />
, DIM5_ID INT<br />
, NUMBER1 FLOAT<br />
, NUMBER2 FLOAT<br />
, NUMBER3 FLOAT<br />
, NUMBER4 FLOAT<br />
) engine=ARCHIVE<br />
;</code></p>
<p>and then rerun your ETL.  Archive tables bear a certain burden, but they can be worthwhile.  Admittedly, I&#8217;ve not heard how this customer did with it operationally so I don&#8217;t know how it worked for them on a day to day basis.  <img src='http://www.nicholasgoodman.com/bt/blog/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: margiex</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-46799</link>
		<dc:creator>margiex</dc:creator>
		<pubDate>Mon, 21 May 2007 14:16:30 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-46799</guid>
		<description>if loaded one day's data which is wrong, how can we reload it? because we can't use delete or update.</description>
		<content:encoded><![CDATA[<p>if loaded one day&#8217;s data which is wrong, how can we reload it? because we can&#8217;t use delete or update.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Matt Casters</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-39844</link>
		<dc:creator>Matt Casters</dc:creator>
		<pubDate>Wed, 18 Apr 2007 20:49:49 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-39844</guid>
		<description>Brian, does it work in combination with table partitioning as well?</description>
		<content:encoded><![CDATA[<p>Brian, does it work in combination with table partitioning as well?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Brian Aker</title>
		<link>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-39824</link>
		<dc:creator>Brian Aker</dc:creator>
		<pubDate>Wed, 18 Apr 2007 17:26:08 +0000</pubDate>
		<guid>http://www.nicholasgoodman.com/bt/blog/2007/04/18/mysql-archive-tablespace-for-facts/#comment-39824</guid>
		<description>Hi!

This is with 5.0 right? In 5.1 Archive's reading got a lot faster and its insertion improved as well (especially with tables that have lots of nulls).  I spent a little time with an 8way machine making sure archive would work with multiple processors well.

Archive does not have update/delete for reasons of market. I was asked not to implement them. I've got it on paper to do it, but its not really high on my list at the moment (the new skip indexes are though, which will make scans faster).

Cheers,
  -Brian</description>
		<content:encoded><![CDATA[<p>Hi!</p>
<p>This is with 5.0 right? In 5.1 Archive&#8217;s reading got a lot faster and its insertion improved as well (especially with tables that have lots of nulls).  I spent a little time with an 8way machine making sure archive would work with multiple processors well.</p>
<p>Archive does not have update/delete for reasons of market. I was asked not to implement them. I&#8217;ve got it on paper to do it, but its not really high on my list at the moment (the new skip indexes are though, which will make scans faster).</p>
<p>Cheers,<br />
  -Brian</p>
]]></content:encoded>
	</item>
</channel>
</rss>
