<?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: Say it Ain&#8217;t So Memcache</title>
	<atom:link href="http://www.nodroidsallowed.com/2009/06/13/say-it-aint-so-memcache/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.nodroidsallowed.com/2009/06/13/say-it-aint-so-memcache/</link>
	<description>Practical techniques for raising a well-adjusted database</description>
	<lastBuildDate>Thu, 10 Jun 2010 01:20:24 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: vdibart</title>
		<link>http://www.nodroidsallowed.com/2009/06/13/say-it-aint-so-memcache/comment-page-1/#comment-1812</link>
		<dc:creator>vdibart</dc:creator>
		<pubDate>Mon, 24 Aug 2009 02:19:41 +0000</pubDate>
		<guid isPermaLink="false">http://www.nodroidsallowed.com/?p=47#comment-1812</guid>
		<description>@Dustin: A little background would help.  I have a bunch of PHP objects that represent stats and detail information for each MLB player - so position name, team name, batting average, stolen bases, etc.  For simplicity&#039;s sake, assume some other process created these objects and stored them for use in various places across the application.  So the comparison was:

1) querying all the player detail information for a random subset of players (including an inner join across a handful of tables), creating and populating objects, then returning them to the view layer

2) querying for just player IDs for a random subset of players (including a very straightforward query with possibly 1 join), then pulling the objects for those player IDs out of memcache.

As I mention, the memory issue was almost as important as the other ones because I would have thought memcache would have reduced memory consumption, but of course that seems naive now.  Those objects have to decompressed, marshalled, whatever, and all that happens in memory.</description>
		<content:encoded><![CDATA[<p>@Dustin: A little background would help.  I have a bunch of PHP objects that represent stats and detail information for each MLB player &#8211; so position name, team name, batting average, stolen bases, etc.  For simplicity&#8217;s sake, assume some other process created these objects and stored them for use in various places across the application.  So the comparison was:</p>
<p>1) querying all the player detail information for a random subset of players (including an inner join across a handful of tables), creating and populating objects, then returning them to the view layer</p>
<p>2) querying for just player IDs for a random subset of players (including a very straightforward query with possibly 1 join), then pulling the objects for those player IDs out of memcache.</p>
<p>As I mention, the memory issue was almost as important as the other ones because I would have thought memcache would have reduced memory consumption, but of course that seems naive now.  Those objects have to decompressed, marshalled, whatever, and all that happens in memory.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: vdibart</title>
		<link>http://www.nodroidsallowed.com/2009/06/13/say-it-aint-so-memcache/comment-page-1/#comment-1811</link>
		<dc:creator>vdibart</dc:creator>
		<pubDate>Mon, 24 Aug 2009 02:09:56 +0000</pubDate>
		<guid isPermaLink="false">http://www.nodroidsallowed.com/?p=47#comment-1811</guid>
		<description>Memcache was running on the web/php slice.  Sorry, should have included that above.</description>
		<content:encoded><![CDATA[<p>Memcache was running on the web/php slice.  Sorry, should have included that above.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steve Yen</title>
		<link>http://www.nodroidsallowed.com/2009/06/13/say-it-aint-so-memcache/comment-page-1/#comment-1803</link>
		<dc:creator>Steve Yen</dc:creator>
		<pubDate>Sat, 22 Aug 2009 20:53:13 +0000</pubDate>
		<guid isPermaLink="false">http://www.nodroidsallowed.com/?p=47#comment-1803</guid>
		<description>Interesting results.  I&#039;m wondering which slice was your memcached server was running?  On your web/php slice or database slice?</description>
		<content:encoded><![CDATA[<p>Interesting results.  I&#8217;m wondering which slice was your memcached server was running?  On your web/php slice or database slice?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dustin Sallings</title>
		<link>http://www.nodroidsallowed.com/2009/06/13/say-it-aint-so-memcache/comment-page-1/#comment-1789</link>
		<dc:creator>Dustin Sallings</dc:creator>
		<pubDate>Sat, 22 Aug 2009 05:52:24 +0000</pubDate>
		<guid isPermaLink="false">http://www.nodroidsallowed.com/?p=47#comment-1789</guid>
		<description>Well, without seeing your code, I can only guess that you&#039;re doing a PK lookup on a table from a DB pool and comparing it to connecting to a memcached server and issuing a single request and disconnecting.

There are places where memcached will help you a lot, and places where it doesn&#039;t provide much value.  You should start by measuring the places where you&#039;re spending a lot of time in the DB, or constructing an object that requires multiple database queries.

(probably won&#039;t see any responses as I typically don&#039;t comment on random blogs since I can&#039;t track easily track responses)</description>
		<content:encoded><![CDATA[<p>Well, without seeing your code, I can only guess that you&#8217;re doing a PK lookup on a table from a DB pool and comparing it to connecting to a memcached server and issuing a single request and disconnecting.</p>
<p>There are places where memcached will help you a lot, and places where it doesn&#8217;t provide much value.  You should start by measuring the places where you&#8217;re spending a lot of time in the DB, or constructing an object that requires multiple database queries.</p>
<p>(probably won&#8217;t see any responses as I typically don&#8217;t comment on random blogs since I can&#8217;t track easily track responses)</p>
]]></content:encoded>
	</item>
</channel>
</rss>

<!-- Dynamic Page Served (once) in 0.280 seconds -->
