Re: [dc] draft-khasnabish-vmmi-problems-00.txt

[Melinda Shore <[e-mail]>]

On 01/20/2012 07:32 AM, Ashish Dalela (adalela) wrote:
> Let's assume we are building protocols for hypervisors.

This may be the source of some of the confusion.  This is
a very vertical framing of the question, when more typically
we're dealing with horizontal communication.   Network
layering is not as clean as we'd necessarily like but what
we're usually dealing with is protocol peers talking
to each other.  That is to say, we're not building protocols
for *hypervisors*, we're building protocols for some functional
component of the network.  From what I've seen a bunch of us
are unclear on why you think that a hypervisor cannot be
where that component lives.

Melinda