"Kastenholz, Frank" <FKastenholz@unispherenetworks.com> writes:
> Is mobile IP inadequate for host mobility, and if so,
> how?
I do not know, but I can try understanding how it evolved.
The simplest Mobile IPv6 requires only the addition of one Home Agent
and modifications of the nodes that move. That's nice.
What came next was the separation wide-area domain/local mobility
domain where Mobile IP can be used for wide-area communication Mobile
IP-like extensions for local domain. This is adding more and more
variations of "agents" and "caches" and it starts ressembling more and
more to what a routing protocol does and thus one might rightfully
question whether Mobile IP was a good start indeed.
The other direction is how Mobile IP might be modified to support
moving prefixes. It looks very logical to require this, since prefix
and address are very tightly related concepts. If Mobile IP can not
be easily adapted to support moving prefixes then this would bring
back again the question: how adequate is Mobile IP for host mobility.
What I'm trying to say is that there's one thing of how adequate is
Mobile IP for host mobility and entirely another thing of how good the
evolutions of Mobile IP are. And I feel that answers to the latter
should influence answers to the first, but I don't know how to express
that.
Alexandru
PS: my oppinions only, and I was not involved in the philosophy behind
Mobile IP. Mobile IP proponents can rightfully disagree with my
text.
This archive was generated by hypermail 2b29 : Mon Aug 04 2003 - 04:10:04 EDT