<?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: A Simple Beowulf Cluster</title>
	<atom:link href="http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/</link>
	<description></description>
	<lastBuildDate>Mon, 06 Feb 2012 08:13:23 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: mike lee</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-191906</link>
		<dc:creator>mike lee</dc:creator>
		<pubDate>Fri, 03 Feb 2012 02:37:40 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-191906</guid>
		<description>hi,  I am a 7th grade math teacher and thought for PI day, it would be fun to try a beowulf cluster to run the computation.  I saw there a was a code download that could do that.  Have you updated any of your instructions, and do you have any words of wisdom for a linux newbee&gt;

thanks</description>
		<content:encoded><![CDATA[<p>hi,  I am a 7th grade math teacher and thought for PI day, it would be fun to try a beowulf cluster to run the computation.  I saw there a was a code download that could do that.  Have you updated any of your instructions, and do you have any words of wisdom for a linux newbee&gt;</p>
<p>thanks</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: puppian</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-89773</link>
		<dc:creator>puppian</dc:creator>
		<pubDate>Sun, 09 Jan 2011 14:25:17 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-89773</guid>
		<description>try puppy linux at http://murga-linux.com/puppy/viewtopic.php?p=484243#484243

you dont need to install anything to try mpich-cluster</description>
		<content:encoded><![CDATA[<p>try puppy linux at <a href="http://murga-linux.com/puppy/viewtopic.php?p=484243#484243" rel="nofollow">http://murga-linux.com/puppy/viewtopic.php?p=484243#484243</a></p>
<p>you dont need to install anything to try mpich-cluster</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dan</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-69573</link>
		<dc:creator>Dan</dc:creator>
		<pubDate>Sun, 03 Oct 2010 14:09:00 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-69573</guid>
		<description>I am wanting to set up a few computers as a cluster. 
Some may be Intel and others AMD. 
A mix of motherboards and CPU Cores and speeds. 
Would it be best to go to high Core Dual Scoket 128gb boards for bettter performance? 

I am looking for the fastest operating system to do a lot of number crunching.

I also want high available cluster.  

Other users may login and do computing or play games over the web. 

I am considering virtual KVM or Single System Image. 

Can you tell me which type KVM or SSI is the best for high performance computing? 
User friendly? 

The biggest limitation is going to be data transfer... Other than a giga bit sec LAN 
Any suggestions with out a large cost? 

Also I was thinking the 6gb/s Sata disk drives would help with this but are about 
double the cost of 3gb/s so I loose about double the storage with the higher cost.

Would some Gpgpu&#039;s be worth the extra cost to get more number crunching done? 
Should they be installed on each node to distribute the work load? 

One other question... If I can get someone to donate computer time (over the web) can 
their computer become a node on my system?
 
How do I do this? 
Any limits to the scaling? 
Will the SSI do the load balancing automatically? 
How would load balancing be accomplished with KVM? 
Can Ubuntu do all this or is there better and faster for a big number crunching system.</description>
		<content:encoded><![CDATA[<p>I am wanting to set up a few computers as a cluster.<br />
Some may be Intel and others AMD.<br />
A mix of motherboards and CPU Cores and speeds.<br />
Would it be best to go to high Core Dual Scoket 128gb boards for bettter performance? </p>
<p>I am looking for the fastest operating system to do a lot of number crunching.</p>
<p>I also want high available cluster.  </p>
<p>Other users may login and do computing or play games over the web. </p>
<p>I am considering virtual KVM or Single System Image. </p>
<p>Can you tell me which type KVM or SSI is the best for high performance computing?<br />
User friendly? </p>
<p>The biggest limitation is going to be data transfer&#8230; Other than a giga bit sec LAN<br />
Any suggestions with out a large cost? </p>
<p>Also I was thinking the 6gb/s Sata disk drives would help with this but are about<br />
double the cost of 3gb/s so I loose about double the storage with the higher cost.</p>
<p>Would some Gpgpu&#8217;s be worth the extra cost to get more number crunching done?<br />
Should they be installed on each node to distribute the work load? </p>
<p>One other question&#8230; If I can get someone to donate computer time (over the web) can<br />
their computer become a node on my system?</p>
<p>How do I do this?<br />
Any limits to the scaling?<br />
Will the SSI do the load balancing automatically?<br />
How would load balancing be accomplished with KVM?<br />
Can Ubuntu do all this or is there better and faster for a big number crunching system.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Anuj Patel</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-38886</link>
		<dc:creator>Anuj Patel</dc:creator>
		<pubDate>Thu, 01 Oct 2009 12:20:55 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-38886</guid>
		<description>hi there,
i have taken up a cluster implementation as a project in college.

problem i am facing is just described below if you can help me out please do so.


and other thing is i am implementing it with MPICH 2.
i have configured MPICH as per guide.

i haven&#039;t configured NIS or NFS.
is it really necessary to configure NIS or NFS?

(connect_lhs 918): failed to connect to lhs at destiny2 45774
localhost.localdomain_33293 (enter_ring 873): lhs connect failed
localhost.localdomain_33293 (run 263): failed to enter ring


Thanks in advance.
Please do reply ASAP.</description>
		<content:encoded><![CDATA[<p>hi there,<br />
i have taken up a cluster implementation as a project in college.</p>
<p>problem i am facing is just described below if you can help me out please do so.</p>
<p>and other thing is i am implementing it with MPICH 2.<br />
i have configured MPICH as per guide.</p>
<p>i haven&#8217;t configured NIS or NFS.<br />
is it really necessary to configure NIS or NFS?</p>
<p>(connect_lhs 918): failed to connect to lhs at destiny2 45774<br />
localhost.localdomain_33293 (enter_ring 873): lhs connect failed<br />
localhost.localdomain_33293 (run 263): failed to enter ring</p>
<p>Thanks in advance.<br />
Please do reply ASAP.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: evad1089</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-29961</link>
		<dc:creator>evad1089</dc:creator>
		<pubDate>Fri, 17 Apr 2009 16:48:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-29961</guid>
		<description>Thank you very much for the tutorial. It was very helpful.

I was wondering if you have any experience with installing the Perl module for MPI? I cannot seem to find much documentation.

Regards,
Dave</description>
		<content:encoded><![CDATA[<p>Thank you very much for the tutorial. It was very helpful.</p>
<p>I was wondering if you have any experience with installing the Perl module for MPI? I cannot seem to find much documentation.</p>
<p>Regards,<br />
Dave</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: kwong</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-27609</link>
		<dc:creator>kwong</dc:creator>
		<pubDate>Fri, 06 Mar 2009 00:33:17 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-27609</guid>
		<description>The short answer is a definitely yes. But unless it is for testing purpose, performance will be the biggest concern when virtualizing cluster notes.</description>
		<content:encoded><![CDATA[<p>The short answer is a definitely yes. But unless it is for testing purpose, performance will be the biggest concern when virtualizing cluster notes.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: WiNeOS</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-27606</link>
		<dc:creator>WiNeOS</dc:creator>
		<pubDate>Thu, 05 Mar 2009 22:36:08 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-27606</guid>
		<description>uhmmm... could it be possible to virtualize the cluster and install a virtual machine that encapsulate all nodes into a big HPC server?</description>
		<content:encoded><![CDATA[<p>uhmmm&#8230; could it be possible to virtualize the cluster and install a virtual machine that encapsulate all nodes into a big HPC server?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jason</title>
		<link>http://www.kerrywong.com/2008/11/04/a-simple-beowulf-cluster/comment-page-1/#comment-25160</link>
		<dc:creator>Jason</dc:creator>
		<pubDate>Sat, 10 Jan 2009 07:24:31 +0000</pubDate>
		<guid isPermaLink="false">http://www.kerrywong.com/?p=404#comment-25160</guid>
		<description>Great info! Looking forward to doing this on Amazon&#039;s EC2.</description>
		<content:encoded><![CDATA[<p>Great info! Looking forward to doing this on Amazon&#8217;s EC2.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

