<?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"
	>
<channel>
	<title>Comments on: Ruby Messaging Shootout</title>
	<atom:link href="http://devver.net/blog/2008/06/ruby-messaging-shootout/feed/" rel="self" type="application/rss+xml" />
	<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/</link>
	<description>A Boulder startup improving the way developers work.</description>
	<pubDate>Fri, 21 Nov 2008 03:48:21 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6.3</generator>
		<item>
		<title>By: Jay Phillips</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-103</link>
		<dc:creator>Jay Phillips</dc:creator>
		<pubDate>Fri, 04 Jul 2008 20:01:58 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-103</guid>
		<description>What about the EventMachine-based Ruby Stomp server? I keep noticing this option gets completely ignored in many Ruby discussions about messaging.

The gem for the server is "stompserver" and the client gem is "stomp". I use it for several production applications.</description>
		<content:encoded><![CDATA[<p>What about the EventMachine-based Ruby Stomp server? I keep noticing this option gets completely ignored in many Ruby discussions about messaging.</p>
<p>The gem for the server is &#8220;stompserver&#8221; and the client gem is &#8220;stomp&#8221;. I use it for several production applications.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dan</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-97</link>
		<dc:creator>Dan</dc:creator>
		<pubDate>Wed, 02 Jul 2008 15:23:40 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-97</guid>
		<description>Yeah, we only posted the SQS results from our home test. Internally on EC2, it is 10x faster, I should have really posted those results as well. I am also hearing that other languages send and receive a lot faster than ruby. I think the default Amazon SQS library does XML handling via Ruby, it would probably be faster to use one of the C xml libraries, I have heard that other SQS libraries are  more robust and better than Amazon's so perhaps I should run another test using Right-AWS or the Active Messaging plugin to get some SQS times. 

Mitch running SQS internally on EC2 our puts were about 0.04 seconds, have you run anything between EC2 instances?

Thanks for all the feedback good to know.</description>
		<content:encoded><![CDATA[<p>Yeah, we only posted the SQS results from our home test. Internally on EC2, it is 10x faster, I should have really posted those results as well. I am also hearing that other languages send and receive a lot faster than ruby. I think the default Amazon SQS library does XML handling via Ruby, it would probably be faster to use one of the C xml libraries, I have heard that other SQS libraries are  more robust and better than Amazon&#8217;s so perhaps I should run another test using Right-AWS or the Active Messaging plugin to get some SQS times. </p>
<p>Mitch running SQS internally on EC2 our puts were about 0.04 seconds, have you run anything between EC2 instances?</p>
<p>Thanks for all the feedback good to know.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mitch Garnaat</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-96</link>
		<dc:creator>Mitch Garnaat</dc:creator>
		<pubDate>Wed, 02 Jul 2008 13:19:56 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-96</guid>
		<description>As I mention here:

http://developer.amazonwebservices.com/connect/thread.jspa?threadID=22892&#38;tstart=0

my put times were closer to 0.08 seconds for small test messages to an existing queue.  This is from Python using boto on a 384K upload link.  SQS will definitely not be the fastest solution but it really shouldn't be as slow as you indicate in your benchmarks.</description>
		<content:encoded><![CDATA[<p>As I mention here:</p>
<p><a href="http://developer.amazonwebservices.com/connect/thread.jspa?threadID=22892&amp;tstart=0" rel="nofollow">http://developer.amazonwebservices.com/connect/thread.jspa?threadID=22892&amp;tstart=0</a></p>
<p>my put times were closer to 0.08 seconds for small test messages to an existing queue.  This is from Python using boto on a 384K upload link.  SQS will definitely not be the fastest solution but it really shouldn&#8217;t be as slow as you indicate in your benchmarks.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Kavanagh</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-91</link>
		<dc:creator>David Kavanagh</dc:creator>
		<pubDate>Wed, 02 Jul 2008 12:21:52 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-91</guid>
		<description>I just did a test, sending 50 messages and the time per send is: 0.28656, and that isn't threaded either, and it's over a DSL line from home. Right away, it's 3x faster and I don't think I'm reproducing your test env properly either.</description>
		<content:encoded><![CDATA[<p>I just did a test, sending 50 messages and the time per send is: 0.28656, and that isn&#8217;t threaded either, and it&#8217;s over a DSL line from home. Right away, it&#8217;s 3x faster and I don&#8217;t think I&#8217;m reproducing your test env properly either.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Kavanagh</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-90</link>
		<dc:creator>David Kavanagh</dc:creator>
		<pubDate>Wed, 02 Jul 2008 12:17:34 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-90</guid>
		<description>I'm not sure about these SQS results. See the bottom of this thread for my SQS receive/delete message times in Java (with a library I wrote)
http://developer.amazonwebservices.com/connect/thread.jspa?messageID=63449&#38;#63449

I've also been doing heavy-duty messaging with SQS in EC2 for a year or more and I think your send times must be off. I have some test send code I'll try running again.</description>
		<content:encoded><![CDATA[<p>I&#8217;m not sure about these SQS results. See the bottom of this thread for my SQS receive/delete message times in Java (with a library I wrote)<br />
<a href="http://developer.amazonwebservices.com/connect/thread.jspa?messageID=63449&amp;#63449" rel="nofollow">http://developer.amazonwebservices.com/connect/thread.jspa?messageID=63449&amp;#63449</a></p>
<p>I&#8217;ve also been doing heavy-duty messaging with SQS in EC2 for a year or more and I think your send times must be off. I have some test send code I&#8217;ll try running again.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dan</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-81</link>
		<dc:creator>Dan</dc:creator>
		<pubDate>Tue, 01 Jul 2008 17:51:41 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-81</guid>
		<description>No not really, I guess finding the std dev for puts and takes, sort of covers the average delay time of performing an action. I am not specifically sure what kind of test you are looking for, I don't think I am thinking about latency in the same way you are. Any thoughts on the kinds of tests you would like to see or what would cover the issues you are wondering about?</description>
		<content:encoded><![CDATA[<p>No not really, I guess finding the std dev for puts and takes, sort of covers the average delay time of performing an action. I am not specifically sure what kind of test you are looking for, I don&#8217;t think I am thinking about latency in the same way you are. Any thoughts on the kinds of tests you would like to see or what would cover the issues you are wondering about?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Henrik Holst</title>
		<link>http://devver.net/blog/2008/06/ruby-messaging-shootout/#comment-76</link>
		<dc:creator>Henrik Holst</dc:creator>
		<pubDate>Tue, 01 Jul 2008 07:43:25 +0000</pubDate>
		<guid isPermaLink="false">http://devver.net/blog/?p=122#comment-76</guid>
		<description>Have you also performed any latency tests?</description>
		<content:encoded><![CDATA[<p>Have you also performed any latency tests?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
