Re: A tunneling proposal

[xxvaf <xxvaf@mfnx.net>]

> Is there a known limit to the number of routes in the global routing
> table? I know bad things started to happen at about 4000 and 10000, but
> obviously those problems have been solved. Routers run just fine with 100,000
> routes at present, and unless I'm mistaken, the most common types of router
> have CPUs and memory that are well below what most of us have in our desk top
> PCs.

The veracity of "Routers run just fine with 100,000 routes..." depends on
how you define "fine".

Modern routers don't fall over with 100,000 routes in them. But initial table
load and BGP convergence times when paths change are both a lot longer than
many would like.

> On top of that, each route takes a LOT of memory: 240 bytes for the routing 
> table and for each peer route in the BGP table in a Cisco.
> 
> If we can both increase memory and decrease the route/BGP table entry size by
> an order of magnitude, we should be able to run with a global routing table
> of 10,000,000 routes. That's about the entire IPv4 space as individual /24's.

The size of individual routing state entries in modern routers has been the
subject of a great deal of optimization over the years. Don't expect to see
it improved by an order of magnitude or even by a factor of two. 

Memory size is not the principal issue; memory speed and routing table update
bandwidth are.

> I think it's worth it to look at this, because with CIDR it is pretty much
> impossible to efficiently route traffic: many locations are hidden behind
> aggregates. Someone in Chicago has to choose between routing traffic for the
> aggregate to the west coast or the east coast, and in either case some
> traffic will have gone in the wrong direction.

If global routing state size resumes a hyper-exponential growth pattern that
exceeds Moore's law, the problem will get worse faster than CPUs increase
in speed. Eliminating CIDR will guarentee hyper-exponential growth - for all
the talk about how CIDR has "failed" in that growth is continuing, the growth
rate would be far, far worse without it - aggregation has successfully
"hidden" at least an order of magnitude of growth.

CIDR-style aggregation is the only technology that has been shown to contain
routing state growth to a manageable level.

Look back at old archives of the IETF and other lists to read some of Noel
Chiappa's and others' writings on the mathematics of network topology and
addressing. If the "addresses" used by routing don't follow the underlying
network topology, excessive state is introduced. For multihoming to really
work, it needs to use topologically-significant addressing. That suggests
that the "multi" in "multihoming" also implies multiple addresses, with
something like SCTP to handle them intelligently.

	--Vince