[rrg] Summary of architectural solution space

[William Herrin]

On Thu, Dec 4, 2008 at 3:37 PM, Brian E Carpenter
<brian.e.carpenter at gmail.com> wrote:
> If we can show that some aspects of renumbering
> are intrinsically impossible to automate, some classes of solution
> will vanish for sure.

Brian,

I've been playing with that question for a while from a clean-slate
perspective. Given a system dependent on map servers and certain
choices about the layer-4 structure, there don't seem to be any LOC
assignment issues that are especially hard to automate, not even in
the core.

On the other hand, renumbering in TCP/UDP networks? Fugeddaboutit.

Regards,
Bill Herrin


-- 
William D. Herrin ................ herrin at dirtside.com  bill at herrin.us
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004