> From: Albert Cabellos <[email protected]>
Hi, thanks for the comments; I'll reply in detail later, but one thing that
caught my eye quickly:
> MPLS-> Although it is a good analogy, I don't think that MPLS is a good
> example given that with LISP we can't stack labels.
??? LISP does use 'stacked' encapsulations (e.g. for a mobile node moving to
a LISP site)?
And the mapping output could be an MPLS header with a 'pre-loaded' label
stack (that's how you do source routing in a label-based system - I'm not
sure if the existing MPLS stuff uses that capability, but eventually in
Nimrod [which was the place that came up with the idea of label stacks, see
RFC-1753] we realized that was how to minimize state setup, by 'pre-loading'
the flow stack at the time the packet is created).
So I'm not sure I understand this comment?
In any case, I'm OK with using some other example - I just used MPLS because
that's one that had been discussed, because there's a lot of MPLS-capable
infrastructure already deployed. Did you have an alternative suggestion? I
can't quickly come up with one that's as good as MPLS.
Noel
_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp
