<?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: Why is my tweekly.fm so late?</title>
	<atom:link href="http://shotbeak.com/2009/10/05/why-is-my-tweekly-fm-so-late/feed/" rel="self" type="application/rss+xml" />
	<link>http://shotbeak.com/2009/10/05/why-is-my-tweekly-fm-so-late/</link>
	<description>Student, aspiring web entrepreneur and musician.</description>
	<lastBuildDate>Sat, 07 Jan 2012 10:25:00 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: shotbeak</title>
		<link>http://shotbeak.com/2009/10/05/why-is-my-tweekly-fm-so-late/comment-page-1/#comment-757</link>
		<dc:creator>shotbeak</dc:creator>
		<pubDate>Mon, 19 Oct 2009 12:29:36 +0000</pubDate>
		<guid isPermaLink="false">http://shotbeak.com/?p=361#comment-757</guid>
		<description>Thanks for the comment. That is one of the ways I&#039;m looking to solve it all. Blog post coming sometime this week detailing what I&#039;m want to do, to fix Tweekly.fm.</description>
		<content:encoded><![CDATA[<p>Thanks for the comment. That is one of the ways I&#39;m looking to solve it all. Blog post coming sometime this week detailing what I&#39;m want to do, to fix Tweekly.fm.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: xpaulbettsx</title>
		<link>http://shotbeak.com/2009/10/05/why-is-my-tweekly-fm-so-late/comment-page-1/#comment-756</link>
		<dc:creator>xpaulbettsx</dc:creator>
		<pubDate>Mon, 19 Oct 2009 05:54:37 +0000</pubDate>
		<guid isPermaLink="false">http://shotbeak.com/?p=361#comment-756</guid>
		<description>Instead of batching all of the tweets up into one giant weekly cronjob, instead randomly pick a day of the week for each user (or better yet, a random time), and evenly bucket them - this way your server is running more often, but it won&#039;t have nearly as many users to deal with on each invocation. It&#039;ll also be way more predictable. &lt;br&gt;&lt;br&gt;Another thing you could do is switch the service to using Amazon EC - this fits much better with your burst model, so that you can reserve 10-20 machines for the update, then drop them all later.</description>
		<content:encoded><![CDATA[<p>Instead of batching all of the tweets up into one giant weekly cronjob, instead randomly pick a day of the week for each user (or better yet, a random time), and evenly bucket them &#8211; this way your server is running more often, but it won&#39;t have nearly as many users to deal with on each invocation. It&#39;ll also be way more predictable. </p>
<p>Another thing you could do is switch the service to using Amazon EC &#8211; this fits much better with your burst model, so that you can reserve 10-20 machines for the update, then drop them all later.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

