Re: [sami] A new draft on state migration use cases is submitted.

[Melinda Shore <[e-mail]>]

On 10/10/11 11:51 PM, Juergen Schoenwaelder wrote:
> Your proposition is that moving filtering rules from the endpoint (the
> hypervisor) into the network is cheaper and easier to scale. To make
> any sense of the above quoted document, I would need to know what the
> performance of middleboxes is going to be and I would need to know how
> many hypervisors are going to be behind a middlebox.

Well, I'm not sure.  It seems to me that "easier to scale" is an issue
only if what's being done fails to scale with the number of rules.  The
cited paper doesn't do that - in fact, it says that if you use freely
available open source software, the marginal cost of adding more rules
is 0.

I agree with your point more generally, however.  It's counterintuitive
to think that it's cheaper to aggregate rulesets in network devices than
it is to push the platform-specific ones out to the edges.  But, unless
there's a real problem here I think the discussion is irrelevant, at
least in the context of trying to figure out whether or not to charter
some IETF work.

Melinda