I'll bet you nickels to doughnuts that it won't make much
of a difference -- in the fact that too may end-ASs originate
specifics to attempt to "engineer" their traffic....
- ferg
I'll bet you nickels to doughnuts that it won't make much
of a difference -- in the fact that too may end-ASs originate
specifics to attempt to "engineer" their traffic....
You could always send those networks a bill for your next BGP-speaking
core upgrade. TCAM entries are expensive y'know.
From what I've seen here and other lists there seems to be a good
financial reason now to aggregate thanks to implementation choices
by vendors and purchasing choices by providers.
Adrian (239k is enough for everyone, or something.)
We got hit by this a couple of months back. We had held out from doing
policy based routing for many many years, but the sheer number of small
routes grew to the point where we couldn't handle the full routing table
and run DCEF at the same time. We were kind of forced into it. Last I
knew, we accepted anything down to a /24 for Arin allocations, and dropped
everything below the minimum allocation size for all the other registries.
So far, it has been working great for us, and we were able to buy another
year or two out of our Cisco gear. But I can see the end of the line
coming and I believe we are going to need to replace our core routers
sooner rather than later.
Sigh....