<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<TITLE>Re: [Beowulf] multi-threading vs. MPI</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<P><FONT SIZE=2>Distributed objects... Corba... Soap... This all precedes multiple cores per server and essentially is message passing in the enterprise.<BR>
<BR>
Michael<BR>
<BR>
Sent from my GoodLink synchronized handheld (www.good.com)<BR>
<BR>
<BR>
-----Original Message-----<BR>
From: Toon Knapen [<A HREF="mailto:toon.knapen@gmail.com">mailto:toon.knapen@gmail.com</A>]<BR>
Sent: Friday, December 07, 2007 01:31 PM Pacific Standard Time<BR>
To: Greg Lindahl<BR>
Cc: beowulf@beowulf.org<BR>
Subject: Re: [Beowulf] multi-threading vs. MPI<BR>
<BR>
Greg Lindahl wrote:<BR>
><BR>
> In real life (i.e. not HPC), everyone uses message passing between<BR>
> nodes. So I don't see what you're getting at.<BR>
><BR>
<BR>
<BR>
Many on this list suggest that using multiple MPI-processes on one and<BR>
the same node is superior to MT approaches IIUC. However I have the<BR>
impression that almost the whole industry is looking into MT to benefit<BR>
from multi-core without even considering message-passing. Why is that so?<BR>
<BR>
toon<BR>
_______________________________________________<BR>
Beowulf mailing list, Beowulf@beowulf.org<BR>
To change your subscription (digest mode or unsubscribe) visit <A HREF="http://www.beowulf.org/mailman/listinfo/beowulf">http://www.beowulf.org/mailman/listinfo/beowulf</A><BR>
</FONT>
</P>
</BODY>
</HTML>