Avri,
> i am not sure i understand exactly what is meant in this
> requirement, so a response to my comment may be yes, that
> fits with what we meant - we do not preclude ...
>
> i figure it is worth discussing anyway.
>
> > 3.19 Media Independence
> >
> > While it is an article of faith that IP operates over a
> > wide variety of media (such as Ethernet, X.25, ATM, and
> > so on), IP routing must take an agnostic view towards
> > any "routing" or "topology" services that are offered
> > by the medium over which IP is operating. That is, the
> > new architecture MUST NOT be designed to integrate
> > with any media-specific topology management or routing
> > scheme.
>
> maybe i have been spending too much time thining about it
> lately, but it seems to me that the entire GMPLS
> architecutre places some constriants on the routing
> architecture that run counter to this requirement.
>
> By saying MUST NOT integrate with media-specifc topology
> management, nrarch-rrg seems to precluding some, if not
> much, of the GMPLS architecutre.
>
> i think it is important that the coming routing architecture
> be able to create the routes used by the underlying media
> layer. this means that the system has to take the native
> capabilites of those layers into account and cannot be
> completely, therfore, independent of them.
>
> at that same time, it may not need to take these mechanisms
> into account when merely doing an IP overlay, but when doing
> routing using the raw structure, it will need to understand
> and be dependent on capabilities, e.g. redundnacy and
> restoration, that exist in many sub layers.
I think you have a valid point. The new architecture should not be
required, but also should not be precluded from integrating with
any media-specific topology management or routing scheme.
Yakov.
This archive was generated by hypermail 2b29 : Mon Aug 04 2003 - 04:10:04 EDT