[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: An idea: GxSE



On Tue, 26 Jun 2001, Ben Black wrote:

> My solution was to distribute "valid" prefixes in the IGP and have them
> communicated to end hosts using an RIP-like mechanism to simply 
> broadcast the set of prefixes hosts on a given segment might be
> translated into (does that make sense?).  Hosts would then inform
> remote ends of sessions (TCP, UDP, what have you), of the prefixes
> from which they might communicate.  

This is similar to the idea I had expressed.

> Given the hop by hop nature of IP, I suspect it will be rather difficult
> to have the translations occur anywhere but at the border.

Not at all.  As I explained in a previous post, having the valid GR/SK
pairs announced in BGP or the like and propagating them through the
internal network via an IGP or a proptocols specifically designed for this
purpose would let you translate in the distribution layer or even access
layer (although in most cases the distribution layer routers would be the
ones beefy enough to handle it of the two).

This moves the intelligence pretty close to the edge.  If you use my idea
of using SK addresses that are globally unique as the actual endpoints of
the connection, your hosts can be dumb as well.  Therefore, it is possible
to do GxSE with multiple remappings where only routers are aware of the
remappings, thus making it backwards compatible with vanilla v6, except
between your border router and the router in your network that is
remapping.  For connections not remapped, it has no effect.

In fact, I think this is very important to note.  It is possible to do
this where remappings are done invisibly to the opposite end if there's
not a GxSE router in the path between your GxSE remapper and the opposite
endpoint of the connection.  While you get a far less effective scenario,
you still get SOME of the benefits, while allowing for vanilla v6.

-Taz

-- 
        "Be liberal in what you accept,
      and conservative in what you send."
--Jon Postel (1943-1998) RFC 1122, October 1989