[ppml] IRTF RRG: finer allocation of IPv4 space

[Robin Whittle]

Denizens of PPML may be interested in joining the IRTF Routing
Research Group list:

  http://www.irtf.org/charter?gtype=rg&group=rrg
  http://psg.com/lists/rrg/2007/maillist.html

The RRG is discussing several ITR-ETR (Ingress Tunnel Router -
Egress Tunnel Router) schemes which are intended primarily to help
the routing system scale well with growing demand for more
multihomed end-user networks - for both IPv4 and IPv6.

I think it would be great to have more hands on deck on the RRG list
to discuss and critique the various proposals.

At some stage, the RRG will recommend to the IETF what new
architectural proposals should be developed further.  I think this
is likely to consist of one of the ITR-ETR schemes (or perhaps some
combination of bits of them) and one or more enhancements to BGP.

All of these ITR-ETR schemes enable address space to be assigned to
end-user networks in finer divisions than is currently practical
with BGP.  Each such division does not involve a BGP advertised
prefix - and in principle, individual IP addresses can be assigned.

This means all these schemes all capable of enabling new management
structures for IPv4 address space, including assigning space in
increments as small as 128, 64, 32 ... 8, 4, 2 or 1 IP addresses.

All these schemes are intended primarily to resolve the problems in
routing scaling, for both IPv4 and IPv6, with incremental deployment
and no changes to hosts or most BGP routers.  While they are all, in
principle, capable of enabling a much more efficient use of IPv4
space, for many more end-user networks, with multihoming and
portability of addresses between ISPs, only my proposal - Ivip - has
this as an explicit goal.

The proposals are:

  LISP-CONS

     http://tools.ietf.org/html/draft-farinacci-lisp-04
     http://tools.ietf.org/html/draft-meyer-lisp-cons-02


  LISP-NERD

     http://tools.ietf.org/html/draft-farinacci-lisp-04
     http://tools.ietf.org/html/draft-lear-lisp-nerd-02

  eFIT-APT

     http://tools.ietf.org/html/draft-jen-apt-00
     http://www.cs.ucla.edu/~lixia/papers/07SIG_IP6WS.pdf
     http://tools.ietf.org/html/draft-wang-ietf-efit-00

  Ivip

     http://www.firstpr.com.au/ip/ivip/
     http://tools.ietf.org/html/draft-whittle-ivip-arch-00

  TRRP

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


Also of interest:

  http://www.firstpr.com.au/ip/ivip/pmtud-frag/
  http://tools.ietf.org/html/draft-templin-inetmtu-05

     Two proposals, applicable to all ITR-ETR schemes,
     for coping with the problems of Path MTU Discovery,
     MTU limits etc. in the tunnel between the ITR and
     the ETR.


  http://www.iab.org/about/workshops/routingandaddressing/
  http://tools.ietf.org/html/rfc4984

     The late 2006 IAB workshop on the routing and addressing
     crisis.


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

     My comparison of LISP-CONS/NERD, eFIT-APT and Ivip.


  http://www.isi.edu/ant/address/
  http://www.isi.edu/~johnh/PAPERS/Heidemann07c.pdf
  http://www.firstpr.com.au/ip/host-density-per-prefix/

     Two recent IPv4 ping surveys find just over 100 million
     ping-responsive IPv4 addresses, which is around 4% of
     advertised addresses.


http://www.ripe.net/ripe/meetings/ripe-55/presentations/huston-ipv4.pdf

     Geoff Huston's "train wreck" presentation, in which he
     estimates that only 5 to 20% of advertised IPv4 address
     space is actually used by a hosts.


http://www.ripe.net/ripe/meetings/ripe-55/presentations/bush-ipv6-transition.pdf

     Randy Bush's presentation on the difficulties of introducing
     IPv6, and how every network using IPv6 will need some IPv4
     space for a long time to come.

 - Robin