I was just thinking (dangerous, I know) and wondering how much extra
paths to a known prefix affected the size of the routing tables under
BGP. More specifically, under Cisco's implementation of BGP.
Multiple paths to a prefix do cause memory consumption, but it is
non-linear (i.e., the memory for two paths for one prefix is less than the
memory for two paths to two different prefixes).
The trend seems to be toward everyone multihoming for 1) stability/fault
tolerance (yes, I know it's only really fault tolerant if you have
separate paths and etc, but thats not the point - the trend is) and 2)
shorter paths (hop counts) to disparate net.places. The hierarchial
model of the net with a few big providers at the top with smaller and
smaller providers underneath until you get to the the base customer is
just not where we're headed (IMHO). I think we're headed more toward a
fully connected graph of AS's (though we may never get there, of course
- probably an asymptotic approach).
An interesting discussion in and of itself, which I will not touch. ;-)
So my question is, how will current routers (and routing technology)
handle this? well? not well?
The routing technology (i.e., BGP) will handle it just fine. The current
routers, will, of course, be overwhelmed. Before you're shocked, please
recall that they will also be overwhelmed even if no one multi-homes. ;-)
If all the ASs in existance were to suddenly peer with each other,
what would happen? would BGP explode? Massive CPU crunch? What
effect would route flaps then have?
Well AS paths will get a lot shorter. ;-) BGP will not explode. But it
will use N times the memory and CPU that it uses now, for N domains.
Recall that BGP propagates only the path actually used by a domain to its
neighbors. The route flaps _would_ have a horrible effect tho. You're
positing an N^2 mesh of AS's. Thus, any site flapping causes each AS to
xmit the change to each other AS. Not pretty, especially if it's not
getting used.
Tony
