RE: cleaned up question sheet

[Jim Boyle <jimpb@nc.rr.com>]

an incomplete set of initial responses to your post.

>
> I think that we need to agree on some more terminology than that mentioned
> in A.1. I had to make some definitions to answer the questions, as noted

I agree that a more complete definitions list would be of value

> > B. Network types and protection modes
> >

So I'm taking away that  you think restoration is important in:

	Yes:
	- TDM (SONET/SDH)
	- data (MPLS)
	- data (IP)
	Not quite now:
	- ring-mesh combos as an explicit point
	- optical (e.g. could be sonet, could be hyperchannel in there -
		lambda)
	- p-cylces (see
www.ee.ualberta.ca/~grover/pdf/CCBR98-Grover-Stamate-pcycle-slides.PDF
		if rob's thinking what I'm thinking - note the patent
		pending)  pretty cool stuff, but I think there's some high
		mileage initial steps to be taken.

on the IP one, are you thinking where IP follows standard IP routing and
is fowarded as IP, but in failure, an MPLS backukp LSP helps out?

> >
> > 2.  What are the requirements with regard to
> >      the protection modes to be supported in each network type covered?
> >      (Examples of protection modes include 1+1, M:N, shared mesh,
> >      UPSR, BLSR, newly defined modes such as P-cycles, etc.)
>
	<snip>
>
> I think that we need to state requirements in terms of the SONET/SDH
> "connection" systems that have protection in Linear or Ring configurations.
> I'm not an expert in this area, but the following type of taxonmy might be
> useful
> - 1:1, 1+1, 1:N Linear Line systems
> - Unidirectional/Bidirectional Line/Path Switched Rings 2/4 wires (6
> combinations)

	while I think there could be some commonality on how one does
protection in TDM and data networks, I see no need to necessarily mimic
anything in the standardized SONET/SDH worlds (e.g. BLSR).  Should there
maybe be some ability to propogate the protection characteristics of links
throughout the network though?

> >
> > C. Hierarchy
> >
> > 1. Vertical (between two network layers):

 	we both noted that layering of protection can be costly.  While
you noted some tuning aspects that should be headed, you did not (nor did
I) call for a comingled control plane.

> > 2. Horizontal (between two areas or administrative subdivisions within
> >     the same network layer):

  you noted that multi-vendor can necessitate horizontal hierarchy (at
least this has been the case in TDM networks).  I see no technical reason
for this though.
  you noted that concatenation of protection approaches can be used to
cacase across horizontal hierarchy.  while not perfect, it's workable.  I
agree.
  you raise an interesing point that if there is failure in one domain,
there is no need to react in another domain.
  on signalling, you raise concern with O(N^2), I see O(N^2) as an
inevitability (though we can argue about what is N, the edge, or
aggregation points).

> >
> > D. Policy
> >
> > 1. What are the requirements for policy support during
> > protection/restoration,
> >     e.g., restoration priority, preemption, etc.
> >

  you noted as I did, that the types of "policy" here are base, and in
relationship to protection/restoration of traffic.  Not elaborate business
application type policy.

>
> > E. Signaling Mechanisms