<?xml version="1.0" encoding="utf-8" ?>
<rss version="2.0" xmlns:osnews="http://osnews.com/rss2#">
	<channel>
		<title>OSNews: </title>
		<link>http://www.osnews.com/story/18640/Serverwide_Performance_Benchmarking</link>
		<description>Exploring the Future of Computing</description>
		<language>en-us</language>
		<copyright>Copyright 2001-2009, David Adams</copyright>
		<webMaster>adam+nospam@osnews.com</webMaster>
		<lastBuildDate>Mon, 09 Nov 2009 07:31:06 GMT</lastBuildDate>
		<image>
			<url>http://www.osnews.com/images/osnews.gif</url>
			<title>OSNews.com</title>
			<link>http://www.osnews.com</link>
		</image>
		<item>
			<title>Missing parameters</title>
			<link>http://osnews.com/thread?272697</link>
			<guid isPermaLink="true">http://osnews.com/thread?272697</guid>
			<description>Readahead and IO scheduler also make a lot of difference in performance. I don't use databases, only desktop, but I found useful performance tips in RedHat, Postgresql, and IBM documents. Right now I'm using JFS with the deadline IO scheduler and I'm impressed. It got even better after setting readahead to 1024. <br />
<br />
Most database performance documents show that the deadline scheduler makes a lot of difference in throughput, and on benchmarks explicity using deadline JFS was almost allways the best. It's not EOL'd as some believe, last release is dated 2007-08-24.<br />
<br />
Try this in any filesystem and you'll see the difference:<br />
echo deadline &gt; /sys/block/sda/queue/scheduler<br />
echo 1024 &gt; /sys/block/sda/queue/nr_requests<br />
echo 250 &gt; /sys/block/sda/queue/iosched/read_expire<br />
<br />
Bruno</description>
			<pubDate>Wed, 19 Sep 2007 23:54:00 GMT</pubDate>
			<author>donotreply@osnews.com (bilu)</author>
			<category>Comments</category>
		</item>
	</channel>
</rss>
