Bill, On 2008-12-24 16:29, William Herrin wrote: > Hi Folks, > > The Summary of Routing Architectures discussed in the RRG, posted at > http://bill.herrin.us/network/rrgarchitectures.html includes: > > Strategy F. > > Do nothing. (RFC 1887 ยง 4.4.1) > > I propose that we REJECT strategy F. > > As we use it now, BGP lacks sufficient scalability to see the Internet > into the coming decades.
Before we take this as revealed truth, let's not forget that BGP4 has in fact scaled amazingly well since 1994. I assert that we still do not have anything like a scientifically adequate projection for the next five or ten years (as we didn't at the Amsterdam IAB workshop). That goes for both projected RIB growth and projected growth in the UPDATE load, and it goes for the engineering of core routers. So we *cannot* exclude that the BGP4 system will continue to scale if we do nothing. I'm not in fact advocating a 'do nothing' strategy. On the contrary, I advocate that the RRG makes the best recommendations that it can. But I suggest that we should neither accept nor reject strategy F; I think we should just set it aside. It's out of our control anyway. Brian > We constrain the growth of the BGP RIB by > disadvantaging small, innovative organizations when they seek IP > addresses from the regional registries. We constrain the growth of the > BGP RIB by refusing to offer consumer level broadband products capable > of multihoming. Then we spend some $2B per year getting the routes > that do make it into BGP to work. [Herrin BGP Cost] This damages the > Internet, damages invention and ultimately damages us. > > I ask you to post either, "I AGREE that we should reject strategy F" > or "I DISAGREE; strategy F should continue to be considered.". I'd > also appreciate a few words about why or why not. > > If we have a strong consensus to reject strategy F, the Summary of > Routing Architectures will be updated to reflect that we have > considered and rejected it. > > Regards, > Bill Herrin > > _______________________________________________ rrg mailing list [email protected] https://www.irtf.org/mailman/listinfo/rrg
