[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: recent slowdown in routing table growth
> > But,as per geoff's presentation, multi-homing/TE will surely
> > be a major contributor in future.
> Why do people keep blaming multihomers for the mess that is the global
> routing table?
Its a prediction and not a proved fact. But, your correct that
address fragmentation and failure to aggregate contribute more to
the growth than multi-homing/load balancing(TE)
> Prefix filtering is evil because it targets the wrong people:
yup. that is a major problem with prefix filtering. we have
to find methods for announcing fine-grained prefixes( to
allow TE/multi-homing.) I don't know if any one have suggestions.
But, http://www.research.att.com/~jrex/papers/filter.pdf suggests
that we can define reserved communities to label those prefixes.
I have some other comments -like how one can cluster a set of
identically announced prefixes to reduce RT size. We can have
it off-list(if you like), as the discussion is far off from