On Mon, Nov 08, 2010 at 09:44:57PM +1100, Justin Clift wrote:
On Nov 8, 2010, at 9:17 PM, Daniel P. Berrange wrote:
<snip>
> I don't think virtual machine is suitable for virsh, because that implies
> full machine virtualization. libvirt/virsh can also be used for containers
> / OS virtualization. Domain/Guest are the only ones that really cover both
> possible usages.
From the technical point of view, yeah. But, terminology for virtualisation seems
to have moved on to such a point that "Virtual Machine" is now a generic term
for any kind of virtualisation tech. The fact that it's "not a full
machine" virtualisation
is largely moot. The different ways we support are just "different approaches"
to
virtual machines (sic).
I don't really agree with this. People using OpenVZ or LXC are simply not
refering to their guests as virtual machines because it makes no sense
as a term in this context.
It's kind of like the conversation the other day about virtual
switches. There's more
than one type around the place (Linux bridging, the openvswitch guys, etc). But,
for the purpose of writing and communicating about them, we can just look at them
as all being types of "virtual switches", albeit with different approaches and
different properties.
Same thing here. "Virtual Machine" is just the generic term now, and is very
widely understood.
No, not really. This is a completely different scenario. You haven't got two
fundamentally different concepts here, this is just a Xen vs KVM vs VMWare
scenario where you've just got different impls of the same concept.
So, I still reckon we should go ahead with this. :)
I don't agree.
Regards,
Daniel
--
|: Red Hat, Engineering, London -o-
http://people.redhat.com/berrange/ :|
|:
http://libvirt.org -o-
http://virt-manager.org -o-
http://deltacloud.org :|
|:
http://autobuild.org -o-
http://search.cpan.org/~danberr/ :|
|: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|