Re: Transport level multihoming

[Ben Black <ben@layer8.net>]

On Tue, Apr 03, 2001 at 01:48:43PM -0500, Brian E Carpenter wrote:
> > I think there is a lot of potential common-intrest here, it would be
> > foolish to ignore it.
> 
> True, but like it or not this will not take away the need for IP level
> multihoming. In fact I think transport layer multihoming would be better
> served by a separate discussion list.
> 

I don't think it is quite yet time for another list.  Although I like the
approach of pushing multihoming into the transport protocol, it has a
fundamental weakness which has to be addressed in general, and not just
with a single protocol like SCTP: Any new transport protocol MUST have
this functionality.  This can be burdensome, particularly for protocols
like RTP that carry around a significant amount of addressing information
for various purposes.  I don't even want to think of the atrocities that
would be committed in the name of global multicast, either.

So, for those supporting the use of transport multihoming, how about
a draft that generalizes the concept?  Protocol Requirements for
Transport Multihoming, perhaps?  In combination with a draft or drafts
giving the previously mentioned transparent API for existing TCP and
UDP apps, this could be viewed as a complete package for one possible
solution to the multihoming problem we face.


Ben