<?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: Cost of Optimization is Greater than the Cost of Synchronization</title>
	<atom:link href="http://www.whirlycott.com/phil/2004/11/23/cost-of-optimization-is-greater-than-the-cost-of-synchronization/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.whirlycott.com/phil/2004/11/23/cost-of-optimization-is-greater-than-the-cost-of-synchronization/</link>
	<description>Philip Jacob's web page</description>
	<lastBuildDate>Tue, 13 Sep 2011 08:59:30 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Mark Boon</title>
		<link>http://www.whirlycott.com/phil/2004/11/23/cost-of-optimization-is-greater-than-the-cost-of-synchronization/comment-page-1/#comment-12</link>
		<dc:creator>Mark Boon</dc:creator>
		<pubDate>Thu, 25 Nov 2004 14:22:09 +0000</pubDate>
		<guid isPermaLink="false">http://www.whirlycott.com/phil/?p=22#comment-12</guid>
		<description>Object pooling can be essential when creating large numbers of objects, be they large or small. What is essential is whether the object can be initialised much faster than it can be allocated. In board-game programming this is widely used, as the program can easily become an order of magnitude slower when doing &#039;new&#039; and let the GC do its work millions of times instead of (just an example) popping a Move element from an ArrayList and set its x and y coordinates and its piece-type.</description>
		<content:encoded><![CDATA[<p>Object pooling can be essential when creating large numbers of objects, be they large or small. What is essential is whether the object can be initialised much faster than it can be allocated. In board-game programming this is widely used, as the program can easily become an order of magnitude slower when doing &#8216;new&#8217; and let the GC do its work millions of times instead of (just an example) popping a Move element from an ArrayList and set its x and y coordinates and its piece-type.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Radu-A. Popescu</title>
		<link>http://www.whirlycott.com/phil/2004/11/23/cost-of-optimization-is-greater-than-the-cost-of-synchronization/comment-page-1/#comment-11</link>
		<dc:creator>Radu-A. Popescu</dc:creator>
		<pubDate>Thu, 25 Nov 2004 08:52:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.whirlycott.com/phil/?p=22#comment-11</guid>
		<description>Er, like I think object pools are generally useful when cost creation is high. Think network, database connections, images etc. Comparing a decent object pool to new Object() is rather pointless.</description>
		<content:encoded><![CDATA[<p>Er, like I think object pools are generally useful when cost creation is high. Think network, database connections, images etc. Comparing a decent object pool to new Object() is rather pointless.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: toto</title>
		<link>http://www.whirlycott.com/phil/2004/11/23/cost-of-optimization-is-greater-than-the-cost-of-synchronization/comment-page-1/#comment-10</link>
		<dc:creator>toto</dc:creator>
		<pubDate>Wed, 24 Nov 2004 12:57:36 +0000</pubDate>
		<guid isPermaLink="false">http://www.whirlycott.com/phil/?p=22#comment-10</guid>
		<description>People doing Object caches are so smart SUN cannot achieve the same things with the JVM ;-) !
Stop this sucker stuff. Just create Objects, and let the JVM handle its job, it is faster and safer. So far Thread pools and Object pool I saw are slow and buggy. The JVM is designed to do the work. Unless you like to reinvent the (same) wheel again and again, go on and do one more Object cache...</description>
		<content:encoded><![CDATA[<p>People doing Object caches are so smart SUN cannot achieve the same things with the JVM <img src='http://www.whirlycott.com/phil/wp-includes/images/smilies/icon_wink.gif' alt=';-)' class='wp-smiley' />  !<br />
Stop this sucker stuff. Just create Objects, and let the JVM handle its job, it is faster and safer. So far Thread pools and Object pool I saw are slow and buggy. The JVM is designed to do the work. Unless you like to reinvent the (same) wheel again and again, go on and do one more Object cache&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bob Lee</title>
		<link>http://www.whirlycott.com/phil/2004/11/23/cost-of-optimization-is-greater-than-the-cost-of-synchronization/comment-page-1/#comment-9</link>
		<dc:creator>Bob Lee</dc:creator>
		<pubDate>Tue, 23 Nov 2004 17:29:09 +0000</pubDate>
		<guid isPermaLink="false">http://www.whirlycott.com/phil/?p=22#comment-9</guid>
		<description>Here&#039;s a faster and simpler approach I use to cache invocation objects in Dynaop (actually I don&#039;t use it anymore--I just create new instances). This approach extends the class it&#039;s pooling, so you could use something like cglib to create a generic implementation.

https://dynaop.dev.java.net/source/browse/dynaop/src/dynaop/PooledInvocation.java?rev=1.6&amp;content-type=text/vnd.viewcvs-markup</description>
		<content:encoded><![CDATA[<p>Here&#8217;s a faster and simpler approach I use to cache invocation objects in Dynaop (actually I don&#8217;t use it anymore&#8211;I just create new instances). This approach extends the class it&#8217;s pooling, so you could use something like cglib to create a generic implementation.</p>
<p><a href="https://dynaop.dev.java.net/source/browse/dynaop/src/dynaop/PooledInvocation.java?rev=1.6&#038;content-type=text/vnd.viewcvs-markup" rel="nofollow">https://dynaop.dev.java.net/source/browse/dynaop/src/dynaop/PooledInvocation.java?rev=1.6&#038;content-type=text/vnd.viewcvs-markup</a></p>
]]></content:encoded>
	</item>
</channel>
</rss>

