Re: Recommendations
pogson <[email protected]> Wed, 18 Jul 2007 10:24:01 -0500
| Newsgroups | gmane.linux.cluster.openmosix.general |
|---|---|
| Message-ID | <1184772241.10402.106.camel@Beast> |
--===============1194339640==
Content-type: multipart/related; type="multipart/alternative";
boundary="=-wWC8+XS5+VTLsx1spSb7"
--=-wWC8+XS5+VTLsx1spSb7
Content-Type: multipart/alternative; boundary="=-EJpsqqCoSTUUbaJHM7OO"
--=-EJpsqqCoSTUUbaJHM7OO
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
On Wed, 2007-18-07 at 07:24 -0700, Wes Wagner wrote:
> The other virtue of
> openmosix is the ability to PXE boot/CD boot random systems all over
> the
> network and borrow them, plus later gracefully shut them down.
Last year I set up an LTSP system that used multiple servers. OpenMosix
showed promise with a very flexible clustering solution. I was wishing
2.6/Opteron/MigSHM were available. As it turned out, I used static
allocation which would be painful whenever I needed to do any work on a
server. Fortunately uptime was good.
I often work in computer labs where 24 PCs sit idling with a few users
clicking and gawking. OM can put them to work while SMP cannot.
The question about the need for OpenMosix vs SMP is interesting. In the
universe there must be quite a mix of needs. Some mixes worked very well
with OM 2.4 . Perhaps the obstacle is "not SMP-safe". As many CPUs are
SMP, it would be very inefficient if the logic of OM could not be made
to work with SMP. Why is OM not SMP-safe if most of OM is now to be in
userland?
There are hundreds of millions of PCs. I would be very surprised if
there were only a few thousand OM clusters out there. Sourceforge shows
384,668 downloads and there are several distros incorporating OM
technology. I am sure activity would increase if 2.6 could be finished.
Maybe it is time I learned C and kernel-hacking. If only I weren't so
old... ;-)
Robert Pogson
--
A problem is an opportunity.
--=-EJpsqqCoSTUUbaJHM7OO
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: 7bit
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.12.1">
</HEAD>
<BODY>
On Wed, 2007-18-07 at 07:24 -0700, Wes Wagner wrote:<BR>
<BLOCKQUOTE TYPE=CITE>
<TT><FONT COLOR="#000000">The other virtue of</FONT></TT><BR>
<TT><FONT COLOR="#000000">openmosix is the ability to PXE boot/CD boot random systems all over the</FONT></TT><BR>
<TT><FONT COLOR="#000000">network and borrow them, plus later gracefully shut them down.</FONT></TT><FONT COLOR="#000000"> </FONT><BR>
</BLOCKQUOTE>
<BR>
Last year I set up an LTSP system that used multiple servers. OpenMosix showed promise with a very flexible clustering solution. I was wishing 2.6/Opteron/MigSHM were available. As it turned out, I used static allocation which would be painful whenever I needed to do any work on a server. Fortunately uptime was good.<BR>
<BR>
I often work in computer labs where 24 PCs sit idling with a few users clicking and gawking. OM can put them to work while SMP cannot.<BR>
<BR>
The question about the need for OpenMosix vs SMP is interesting. In the universe there must be quite a mix of needs. Some mixes worked very well with OM 2.4 . Perhaps the obstacle is "not SMP-safe". As many CPUs are SMP, it would be very inefficient if the logic of OM could not be made to work with SMP. Why is OM not SMP-safe if most of OM is now to be in userland?<BR>
<BR>
There are hundreds of millions of PCs. I would be very surprised if there were only a few thousand OM clusters out there. Sourceforge shows 384,668 downloads and there are several distros incorporating OM technology. I am sure activity would increase if 2.6 could be finished.<BR>
<BR>
Maybe it is time I learned C and kernel-hacking. If only I weren't so old... <IMG SRC="cid:1184772184.10402.104.camel@Beast" ALIGN="middle" ALT=";-)" BORDER="0"><BR>
<BR>
Robert Pogson<BR>
<TABLE CELLSPACING="0" CELLPADDING="0" WIDTH="100%">
<TR>
<TD>
<PRE>
--
A problem is an opportunity.
</PRE>
</TD>
</TR>
</TABLE>
</BODY>
</HTML>
--=-EJpsqqCoSTUUbaJHM7OO--
--=-wWC8+XS5+VTLsx1spSb7
Content-ID: <1184772184.10402.104.camel@Beast>
Content-Disposition: attachment; filename=smiley-4.png
Content-Type: image/png; name=smiley-4.png
Content-Transfer-Encoding: base64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--=-wWC8+XS5+VTLsx1spSb7--
--===============1194339640==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
--===============1194339640==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
_______________________________________________
openMosix-general mailing list
openMosix-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/openmosix-general
--===============1194339640==--