Comments on draft-green-cdnp-gen-arch-01

["Doug Potter" <dougpott@cisco.com>]

My apologies for the late comments...

Sections 2, 3.3, and 3.4 all include a "normalized canonical name space" as
a requirement for CDN peering. I can convince myself that it might be useful
in some models, and maybe even desirable, but I'm having trouble making it a
requirement.

Section 3.2 correctly states that "Direction systems explicitly peer but do
not have interior knowledge of surrogates from other CDNs". However, in
section 2, the "system architecture is a system of systems" implies (in #3 &
#4) direct knowledge of surrogates by the direction systems.

Also in section 2, system #4 states "The DIRECTION PEERING system directs a
REQUEST from a CLIENT to a suitable SURROGATE in a peering CDN". I don't
think a DIRECTION PEERING system directs anything. It simply aggregates and
exchanges DIRECTION information with peering CDNs. The DIRECTION SYSTEMS
within the individual CDNs direct REQUESTs.

Section 3.5.1 states:
  "We assume in this example that the CDN DIRECTION SYSTEM
  R2 has a INTER-CDN peering relationship with the
  AUTHORITATIVE DIRECTION SYSTEM R1 and has informed R1 via
  a peering protocol, similar to BGP but modified for content
  routing information, and that some set of addresses
  including the address of the CLIENT, is in the
  "redirection set" of R2".
I think this statement opens the door on the protocol debate that we're
probably not ready for at this point. Certainly R2 also needs to indicate
it's willingness to service requests for the specific URI and probably a
bunch of other stuff. I'd be inclined to simplify the text to
  "We assume in this example that the CDN DIRECTION SYSTEM
  R2 has a INTER-CDN peering relationship with the
  AUTHORITATIVE DIRECTION SYSTEM R1 and has informed R1 via
  a peering protocol of its willingness to service REQUESTS
  for a set of URIs'.

I am confused by section 4's use of DISTRIBUTION PEERING systems and
DISTRIBUTION CPGs. What is their relationship? Are they the same thing?

Also, the figure in section 3.1 identifies a DIRECTION CPG. How does this
relate to a DIRECTION PEERING system?

My biggest concern with the document is its lack of specifics related to the
actual information shared between 'peers'. Section 4 does a pretty good job
of identifying some specifics of replication, signaling, and advertising for
DISTRIBUTION PEERING. We probably need to get at least this deep in the
areas of DIRECTION and ACCOUNTING PEERING.

drp