[rrg] Proposals which match rrgarchitectures.html

[Robin Whittle]

Here is my attempt to state which proposals match the various parts
of Bill's page (Draft 6):

  http://bill.herrin.us/network/rrgarchitectures.html

I have no idea how to match the Strategy B variants to various
proposals, so I hope someone will clarify this.  Likewise D and E.
Please suggest corrections and I will write up a final version at:

  http://www.firstpr.com.au/ip/ivip/rrgarch/

   - Robin


Strategy A

   These are core-edge separation schemes, with the new functionality
   being implemented in the network (in or near DFZ, ISP and end-user
   CE routers, but also for internal routers for Ivip forwarding
   approaches).  Will work with existing hosts (except that Ivip
   forwarding and PMTUD management won't accept fragmentable
   packets longer than a certain size.)

   LISP in all its variants.
   APT.
   Ivip.
   TRRP.
   Six/One Router (for IPv6 only).

   A paper arguing for this class of solutions (core-edge separation)
   and against the "elimination" class (Strategy B) is:

      Towards a Future Internet Architecture: Arguments for
      Separating Edges from Transit Core
      Dan Jen, Lixia Zhang, Lan Wang, Beichuan Zhang
      http://conferences.sigcomm.org/hotnets/2008/papers/18.pdf


   A1a, A1b, A1c.

      I am not sure about these.  Bill and I are discussing them:

        http://www.irtf.org/pipermail/rrg/2008-December/000591.html

      I also discuss potential improvements to the parts which
      describe Ivip's forwarding approaches.

   A2a   LISP-NERD.

   A2b   Ivip and APT are covered by this, but it is not a complete
         description of either.

   A2c   LISP-ALT and TRRP.

         Where is Six/One Router covered?  It doesn't have a specific
         mapping distribution system, but I guess A2c is a likely
         approach.



   A3a   LISP, APT, TRRP and Six/One Router.

   A3b   Ivip.



   A4a   No proposals I can think of match this.

   A4b   Likewise.

   A4c   I am not sure, but this might apply to the PMTUD approaches
         of LISP, APT and TRRP.  AFAIK, LISP has no such approach.
         I am not sure how APT or TRRP will handle it.

         Six-One router and Ivip's forwarding modes have no PMTUD
         problems.

   A4d  Ivip.

   A4e  Six/One Router (IPv6 only).

   A4f  Ivip's Prefix Label Forwarding (PLF) - for IPv6.

   A4g  Ivip's ETR Address Forwarding (EAF) - for IPv4.



   A5a  AFAIK, this applies to all of this class of proposal:
        LISP, APT, Ivip, TRRP and Six/One Router.

   A5b  } I don't know of any proposal which matches these.
   A5c  }



Strategy B

   This is (roughly, exactly?) the elimination class of proposals.

   This requires new host functions, including AFAIK, changes to
   host stacks, APIs and applications.

   No-one is suggesting such a solution for IPv4 - they are all
   modifications to IPv6, I think.

   ILNP is one such proposal, I think.

   HIP may be another.

   However neither of these is fully documented as a solution to the
   routing scaling problem.

   Can anyone confirm this or suggest other proposals?

   Can anyone write about B1x and B2x with respect to actual
   proposals?


Strategy C

   Geographic aggregation.  Heiner Hummel has been proposing this for
   at least a year - but there is no substantial documentation of
   this or any other such proposal.

   Compact routing perhaps?  See critique:

     On Compact Routing for the Internet
     Dmitri Krioukov, kc claffy, Kevin Fall, Arthur Brady
     ACM SIGCOMM CCR, v.37, n.3, p.41-52, 2007
     http://arxiv.org/abs/0708.2309


Strategy D

   FIB compression for DFZ routers.

   What proposals match this?


Strategy E

   Billing system for advertising each prefix in the DFZ.

   Have there been any such proposals?  AFAIK, none have been
   suggested seriously in RAWS, the RAM list or the RRG.



Strategy F

   NOP.



Strategy G

   The original vision of IPv6, with hope that those who wanted to
   multihome would get two or more ISPs and use SHIM6.  (But this
   is IPv-6 only, offers network operators no obvious network-centric
   controls - and only works if the other host has SHIM6 too.)