Re: SMP-safe

pogson <[email protected]> Wed, 22 Aug 2007 14:41:43 -0500
Newsgroups	gmane.linux.cluster.openmosix.general
Message-ID	<1187811703.6119.3.camel@Beast>
--===============1053901234==
Content-type: multipart/alternative; boundary="=-6/axHWxwmOHAR3lV2Yx6"


--=-6/axHWxwmOHAR3lV2Yx6
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

I have been reading about KVM lately. There are parallels with
OpenMosix: migration and emphasis on user-space tools. Since KVM is SMP
safe but runs only on AMD64 with SVM or Intel with VT I am wondering
whether it would be more useful to provide OM-like tools for KVM instead
of re-inventing the wheel. Clearly OM is delightfully light and migrates
processes efficiently but systems that need that can run on 2.4/32bit.
Perhaps OM for 2.6/64bit should be based on KVM with a somewhat
different tool-set.

I am thinking such a split/fork/versioning could get the end-result
quicker. KVM is already in the kernel, well supported. Let them do the
kernel-surgery and OM concentrate on the user-space tools.

If we have a project that needs to migrate processes we can run 32bit on
AMD64 with 2.4. If we need to migrate whole systems we can run 64bit on
AMD64 with 2.6/KVM. OM 32bit can even run as a guest on KVM. The
possibilities are great.

Robert Pogson

--=-6/axHWxwmOHAR3lV2Yx6
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>
I have been reading about KVM lately. There are parallels with OpenMosix: migration and emphasis on user-space tools. Since KVM is SMP safe but runs only on AMD64 with SVM or Intel with VT I am wondering whether it would be more useful to provide OM-like tools for KVM instead of re-inventing the wheel. Clearly OM is delightfully light and migrates processes efficiently but systems that need that can run on 2.4/32bit. Perhaps OM for 2.6/64bit should be based on KVM with a somewhat different tool-set.<BR>
<BR>
I am thinking such a split/fork/versioning could get the end-result quicker. KVM is already in the kernel, well supported. Let them do the kernel-surgery and OM concentrate on the user-space tools.<BR>
<BR>
If we have a project that needs to migrate processes we can run 32bit on AMD64 with 2.4. If we need to migrate whole systems we can run 64bit on AMD64 with 2.6/KVM. OM 32bit can even run as a guest on KVM. The possibilities are great.<BR>
<BR>
Robert Pogson
</BODY>
</HTML>

--=-6/axHWxwmOHAR3lV2Yx6--



--===============1053901234==
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: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
--===============1053901234==
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

--===============1053901234==--