[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RRG] On the Transitionability of LISP



	Joel/Christian,

On Fri, Aug 03, 2007 at 12:38:15PM -0400, Joel M. Halpern wrote:
> At 12:27 PM 8/3/2007, Christian Vogt wrote:
> >I do agree that, to remain reachable, upgraded edge networks would have
> >to use their old locator space in addition to the new ID space.  What I
> >am concerned about are the following two problems that arise from this:
> >
> >  1. It defeats all of the benefits of the ID/locator split.
> >  2. It makes reliable address resolution infeasible.
> >
> >Regarding problem 1:  The four main benefits envisioned for an
> >ID/locator split are the following.
> >
> >  (i)   Enabling edge networks to route packets via arbitrary providers.
> >  (ii)  Reducing the network reconfiguration cost related to rehoming.
> >  (iii) Reducing the size of the global routing table.
> >  (iv)  Reducing the update frequency of the global routing table.

	All goodness. I would add that there are also a bunch of
	well-known (and other) security benefits that accrue from
	a ID/loc split. 

> >Now, if an edge network is forced to maintain its old locator space in
> >addition to new ID space, then /none/ of (i) through (iv) will be
> >satisfied.  On the contrary, network administration overhead will be
> >increased, and the global routing table will become larger.
> 
> Actually, there seem to be quite a few stages between
>     advertise all locators in BGP, exactly as today
>     advertise nothing in BGP
> 
> For example, at a later stage in deployment one could easily imagine 
> advertising only heavily aggregated reachability in BGP, for those 
> sites that have not upgraded, while using LISP (or other solutions) 
> for the bulk of ones traffic. That would give significant
> benefit without losing connectivity from  the non-upgraded world.

	Right. In addition, these edge networks might not initially go to
	something like LISP; rather someone might provide an ITR
	for them (seems like a service someone might like to provide).

	BTW, I'm not really sure what an edge network is, or what
	such a distinction buys one (especially given that in
	practice [implementation], the distinction between core and
	edge is rapidly disappearing).

> >Regarding problem 2:  A coexistence of locators and IDs implies that the
> >result of address resolution depends on the location of the resolving
> >host:  Hosts in legacy edge networks MUST obtain legacy locators, while
> >hosts in upgraded edge networks SHOULD obtain IDs.
> 
> This is actually somewhat harder.  If hosts actually checked 
> reachability, or really used all the A records they got back, it 
> might suffice as part of transition to use the EID plus aggregatable 
> A information.  (It's too bad DNS A records do not have preferences 
> like MX records, but we can't change that now.)

	Agreed.

	BTW, I explicitly tried not to introduce the term "legacy
	locator" which is why I quoted it in my previous
	message. Let's not go there.

	Dave


Attachment: signature.asc
Description: Digital signature