RFC 1955 (rfc1955) - Page 2 of 5


New Scheme for Internet Routing and Addressing (ENCAPS) for IPNG



Alternative Format: Original Text Document



RFC 1955                       IP Encaps                       June 1996


   This scheme is based on the ideas presented by Deborah Estrin (route
   on ADs), Martha Steenstrup (encapsulation), and probably steals from
   ideas put forward by Noel Chiappa, Van Jacobson , Ross Callon, Dave
   Oran, and everyone else in the ROAD group.

CONTEXT

   I think that we (the ROAD group) agree that in the short term we need
   to make better use of the IP address space.  I think we also (mostly)
   agree that in the long term we need a solution that can deal with a
   very large number of end points and routes, as well as support new
   services such as guarantees of service, source selected routes, etc.
   We do not agree on any of the details of this but do agree that we
   can not figure out a long term solution before March.  We do agree
   that we should start working on a long term solution(s).

   What this leaves is the need for a good medium term solution which
   can keep the Internet going until we can design and deploy a long
   term solution.  The medium term solution wants to be the most "cost
   effective".  It should buy us the most time to develop a long term
   solution and do it with as little change to the existing Internet as
   possible.

   I propose this scheme as a new medium term solution.

NEW SCHEME

   The basic idea is that inter-domain routing be done by routing on
   autonomous domains (AD).  The key is how this is done.  The mechanism
   to do this is for the border routers to encapsulate the original IP
   datagrams with another IP header.  The source and destination
   addresses in the new header (I will call it the AD-Header from here
   on) represent the source and destination ADs.

   When the first (entrance) border router receives a datagram from a
   host or router without an AD-Header it looks at the source and
   destination address and does a DNS lookup to get the addresses for
   the AD-Header.  It then adds an AD-Header and forwards the
   encapsulated datagram to its proper destination AD.

   The border routers would compute AD routes by running a routing
   protocol between themselves.  BGP or even IS-IS or OSPF for that
   matter, would work fine.  As you will see later, they might even be
   better.

   The addresses I propose to use for the AD addresses are plain old IP
   addresses.  A small number of Class A and Class B addresses would be
   reserved for this purpose.  The network number of the address would



Hinden                       Informational