I propose to reject every strategy which immediately depends on a repair system :-). In einer eMail vom 29.12.2008 12:25:09 Westeuropäische Normalzeit schreibt [email protected]:
Strategy C Despite a decade or two to research such things, no-one has come close to devising a routing system which could serve the needs of the interdomain routing system, as BGP does today, and which could scale to handling the number of prefixes we need to solve the routing scaling problem. Then start to think. I have send out many mails and no one responded (unless off-list) and tried to really deal with the issues. Although we have no clear target, I think there would be consensus that the scalable routing solution needs to provide multihoming, TE (and I think "portability) to at least several hundred million end-user networks. No one responded when I said in spite of 300 000 routes two thirds are not even considered. Wrt to some destination each router has three sets of adjacent links: those to closer, those to equidistant, those to more remote neighbor nodes. I cater for evaluating properly all three sets. My accusation that by DV ( BGP ) only one set can be considered was accepted without any objection. Furthermore TARA is the only archtecture which can support multihoming such that both the hosts' interests as well as the networks' interests can be supported. Furthermore TARA has no problem at all if there where several hundred million end-user networks. I gave enough proof: see the road network, see the postal service network, see the railway network. Robin, you was silent too, when I said that It would be absurd to introduce user reachability dissemination as to run railway service right. Since there is no sign of a routing system which could scale to this number of advertised prefixes - including by souping up BGP in some way - I suggest we should form a strong consensus to reject this strategy. If you doesn't want to see the signs, I cannot help you. Further arguments include there being no chance we could introduce a new routing system without prohibitive disruption, and that some of the most discussed alternatives (geographic aggregation) do not respect the business needs of ISPs. One of the LISP folks pointed me to a critique of "compact routing": On Compact Routing for the Internet Dmitri Krioukov, kc claffy, Kevin Fall, Arthur Brady ACM SIGCOMM CCR, v.37, n.3, p.41-52, 2007 I am very familiar with these Compact Routing documents by these authors. Ask them whether they know Kurt Mehlhorns algorithm, published in 1988 ! I am sure they don't. I do not doubt that their observation is correct concerning THEIR understanding of hierarchical routing. But this is not MY understanding of hierarchical routing. Ask them why they think that ANY hierachical network would have stretch = 3 ! Ask them what is the best number of hierarchical levels and also why ! I am sure, their answers will unveil that they don't even know what is the real issue. which may be relevant in deciding to reject Strategy C. Heiner
_______________________________________________ rrg mailing list [email protected] https://www.irtf.org/mailman/listinfo/rrg
