Re: [Ospf-wireless-design] Design team outbrief (draft slides) posted

[Richard Ogier <rich.ogier at earthlink.net>]

> Since you'll not be there, I'll take your post above as a preference to
> continue to study both approaches for another meeting cycle.

Actually, my preference is to focus on the MDR approach while
evaluating possible enhancements to the MDR approach using MPRs
(e.g., selecting MDRs by pruning MPRs and defining LSAs based
on MPRs), a variation of Smart Peering to reduce the number of
adjacencies (although I doubt that the improvement in efficiency
will be worth the additional complexity), and other enhancements
to the MDR approach.

I believe that ORs with Smart Peering can result in a performance
similar to MDRs (after both are implemented in the best possible
way), but in this case the approach that results in the most
natural extension to OSPF (minimizing changes and keeping the
same philosophy) should be preferred.  After all, we are extending
OSPF, not developing a completely new protocol.

Also, the Smart Peering approach was introduced late in the game
and I would not want its consideration to cause even more delays
in achieving our ultimate goal.

My second choice would be to continue to study both approaches,
but preferably in a manner that gets us closer to an acceptable
design.  (For example, I don't think it would be good to have
another surprise just before the next meeting.)

Richard




_______________________________________________
Ospf-wireless-design mailing list
Ospf-wireless-design at lists.ietf.org
https://www1.ietf.org/mailman/listinfo/ospf-wireless-design