Tracy Reed <[email protected]> writes:

> On Tue, Oct 06, 2009 at 03:36:11PM -0400, Luke S Crawford spake thusly:
> > live migration (and failover) both require shared storage.  (failover
> > requires doubling your other compute resources as well)  which usually
> 
> Not necessarily. I have failover and zero SPOF with a cluster of 5
> machines doing virtualization with a hot spare worth of extra
> capacity. The more machines you run in the cluster the lower your
> overall percentage of unused capacity (economies of scale apply).

how does that work?  project remus essentially has you 'live migrating'
to another box constantly, but then never starts that second box
unless/until the first box goes down.  obviously, if you do this, to 
backup one image with 2G ram, you need another image with 2g ram...
no economies of scale there.  I don't know how vmware does it, but I 
guess I'm having a hard time seeing how you can instantly switch to new
hardware from hardware you can't copy off of without alrealdy having copied
the ram off of it.

Now, if you mean just 'when a domain crashes, I start it up on my 
spare hardware'  rather than 'the domain keeps running even if 
the hardware it was running on fails completely'  then you aren't winning 
anything with virtualization. you can do the same thing with hardware 
if you boot from some sort of reliable shared storage.  


_______________________________________________
Discuss mailing list
[email protected]
http://lopsa.org/cgi-bin/mailman/listinfo/discuss
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to