<?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: Ruby scaling up to multiple CPUs</title>
	<atom:link href="http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/feed/" rel="self" type="application/rss+xml" />
	<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/</link>
	<description>Random tech jargon</description>
	<lastBuildDate>Mon, 06 Sep 2010 09:29:35 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
	<item>
		<title>By: abg</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-158</link>
		<dc:creator>abg</dc:creator>
		<pubDate>Tue, 15 Dec 2009 08:35:18 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-158</guid>
		<description>&quot;@another David, 1.9 isn’t useless in Rails today, it works perfectly (all tests pass)…the problems arise in 3rd party gems, which may not be compatible.&quot;

Then their tests are crap. ruby-1.9 &amp; rails has a year-old show-stopping encoding bug which the ruby guys seem to be incompetent to fix. 

See https://rails.lighthouseapp.com/projects/8994/tickets/2188-i18n-fails-with-multibyte-strings-in-ruby-19-similar-to-2038</description>
		<content:encoded><![CDATA[<p>&#8220;@another David, 1.9 isn’t useless in Rails today, it works perfectly (all tests pass)…the problems arise in 3rd party gems, which may not be compatible.&#8221;</p>
<p>Then their tests are crap. ruby-1.9 &amp; rails has a year-old show-stopping encoding bug which the ruby guys seem to be incompetent to fix. </p>
<p>See <a href="https://rails.lighthouseapp.com/projects/8994/tickets/2188-i18n-fails-with-multibyte-strings-in-ruby-19-similar-to-2038" rel="nofollow">https://rails.lighthouseapp.com/projects/8994/tickets/2188-i18n-fails-with-multibyte-strings-in-ruby-19-similar-to-2038</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: roger</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-96</link>
		<dc:creator>roger</dc:creator>
		<pubDate>Sun, 12 Jul 2009 04:16:21 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-96</guid>
		<description>Compare it with multi process KRI? :)
[oh and 1.9.1 is ok with rails]
=r</description>
		<content:encoded><![CDATA[<p>Compare it with multi process KRI? <img src='http://letsgetdugg.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /><br />
[oh and 1.9.1 is ok with rails]<br />
=r</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: victori</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-26</link>
		<dc:creator>victori</dc:creator>
		<pubDate>Thu, 30 Apr 2009 21:17:28 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-26</guid>
		<description>Added Ruby 1.9.1 for the hell of it. Ruby 1.9.1 was compiled with GCC 4.3.2, could not get it to compile with SunCC</description>
		<content:encoded><![CDATA[<p>Added Ruby 1.9.1 for the hell of it. Ruby 1.9.1 was compiled with GCC 4.3.2, could not get it to compile with SunCC</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jimmy</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-19</link>
		<dc:creator>Jimmy</dc:creator>
		<pubDate>Wed, 29 Apr 2009 21:33:11 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-19</guid>
		<description>@snuxoll 
I think Java has had native threads since 1.2 which is around 8 years ago .. hardly recently.</description>
		<content:encoded><![CDATA[<p>@snuxoll<br />
I think Java has had native threads since 1.2 which is around 8 years ago .. hardly recently.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: victori</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-17</link>
		<dc:creator>victori</dc:creator>
		<pubDate>Wed, 29 Apr 2009 17:56:02 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-17</guid>
		<description>@Charles Nutter 

--client performed better since I gave it no warm up time. So I just rolled with that. 

@Jon

Consider that part of the benchmark. I am well aware I am initiating a new object each iteration. Everything is an object in ruby ;-)</description>
		<content:encoded><![CDATA[<p>@Charles Nutter </p>
<p>&#8211;client performed better since I gave it no warm up time. So I just rolled with that. </p>
<p>@Jon</p>
<p>Consider that part of the benchmark. I am well aware I am initiating a new object each iteration. Everything is an object in ruby <img src='http://letsgetdugg.com/wp-includes/images/smilies/icon_wink.gif' alt=';-)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Charles Nutter</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-16</link>
		<dc:creator>Charles Nutter</dc:creator>
		<pubDate>Wed, 29 Apr 2009 12:24:51 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-16</guid>
		<description>FYI, that&#039;s &quot;dash dash server&quot;...I think the formatting turned it into an emdash.

And feel free to stop by #jruby on FreeNode for more discussion.</description>
		<content:encoded><![CDATA[<p>FYI, that&#8217;s &#8220;dash dash server&#8221;&#8230;I think the formatting turned it into an emdash.</p>
<p>And feel free to stop by #jruby on FreeNode for more discussion.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Charles Nutter</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-15</link>
		<dc:creator>Charles Nutter</dc:creator>
		<pubDate>Wed, 29 Apr 2009 12:15:34 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-15</guid>
		<description>FYI, a couple tips running JRuby:

1. pass --server for best performance; unless you&#039;re running on 64-bit JVM, it runs in &quot;client&quot; mode normally, which is quite a bit slower
2. if you find we&#039;re slower than 1.9 for straight-line perf, please file a bug (or try running longer)
3. JRuby has improved by double-digit percentages in almost every release over the past year. Run the most recent for real comparisons.
4. See http://kenai.com/projects/jruby/pages/Benchmarks for more information on how to benchmark JRuby well.</description>
		<content:encoded><![CDATA[<p>FYI, a couple tips running JRuby:</p>
<p>1. pass &#8211;server for best performance; unless you&#8217;re running on 64-bit JVM, it runs in &#8220;client&#8221; mode normally, which is quite a bit slower<br />
2. if you find we&#8217;re slower than 1.9 for straight-line perf, please file a bug (or try running longer)<br />
3. JRuby has improved by double-digit percentages in almost every release over the past year. Run the most recent for real comparisons.<br />
4. See <a href="http://kenai.com/projects/jruby/pages/Benchmarks" rel="nofollow">http://kenai.com/projects/jruby/pages/Benchmarks</a> for more information on how to benchmark JRuby well.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jon</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-14</link>
		<dc:creator>Jon</dc:creator>
		<pubDate>Wed, 29 Apr 2009 12:15:07 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-14</guid>
		<description>if Time.new - time &gt;= 10

You&#039;re creating a new Time object on every iteration, which as well probably taking longer than the division may well synchronise on some central heap mutex.  If so this would make the results invalid.

It would be a better idea if the main thread signalled or stopped the test threads after waiting for 10s.</description>
		<content:encoded><![CDATA[<p>if Time.new &#8211; time &gt;= 10</p>
<p>You&#8217;re creating a new Time object on every iteration, which as well probably taking longer than the division may well synchronise on some central heap mutex.  If so this would make the results invalid.</p>
<p>It would be a better idea if the main thread signalled or stopped the test threads after waiting for 10s.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Turnbull</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-13</link>
		<dc:creator>David Turnbull</dc:creator>
		<pubDate>Wed, 29 Apr 2009 06:00:19 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-13</guid>
		<description>Indeed, fully understood. My point is just to put it in the perspective we&#039;ll need in the next few months, i.e. 1.9 vs jruby, not 1.8.

Too many Davids here, btw. :)</description>
		<content:encoded><![CDATA[<p>Indeed, fully understood. My point is just to put it in the perspective we&#8217;ll need in the next few months, i.e. 1.9 vs jruby, not 1.8.</p>
<p>Too many Davids here, btw. <img src='http://letsgetdugg.com/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: snuxoll</title>
		<link>http://letsgetdugg.com/2009/04/28/ruby-scaling-up-to-multiple-cpus/comment-page-1/#comment-12</link>
		<dc:creator>snuxoll</dc:creator>
		<pubDate>Wed, 29 Apr 2009 05:15:32 +0000</pubDate>
		<guid isPermaLink="false">http://letsgetdugg.com/?p=106#comment-12</guid>
		<description>Erm, is anyone really suprised by this?  It&#039;s 2009 and people still don&#039;t know that ruby 1.8 uses green threads?  So did Java until recently, actually.

While ruby 1.9 uses native threads, it still uses a Global Interpreter Lock (GIL) for the sake of backwards compatibility for C extension developers, which prevents it from running multiple threads at once.  However, fixing this is on the roadmap for 1.9.</description>
		<content:encoded><![CDATA[<p>Erm, is anyone really suprised by this?  It&#8217;s 2009 and people still don&#8217;t know that ruby 1.8 uses green threads?  So did Java until recently, actually.</p>
<p>While ruby 1.9 uses native threads, it still uses a Global Interpreter Lock (GIL) for the sake of backwards compatibility for C extension developers, which prevents it from running multiple threads at once.  However, fixing this is on the roadmap for 1.9.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
