On 21/10/2014 19:22, John G. Scudder wrote: > where where in the > the spring
cut-n-paste typo. Not present in the .txt. > I agree with all that. However I think the point at which DFZ size > becomes the upper bound is well below the point at which a practical > problem rears its ugly head. the reason the draft was originally written was because scaling problems were hurting production deployments. Many of these problems are handled better in the more sophisticated RS implementations, but we got a lot of feedback from several RS developers who were surprised at how subtly difficult optimisation was in practice. From this perspective, the authors felt that it was important for RS operators to understand that there's more going on under the hood of a RS than they might realise, so that they won't go off and do something that might on some level seem like a good idea, but which would inadvertently pessimize away any algorithm efficiencies in modern RS implementations. Also there are some more elderly RS implementations which still suffer from many of the problems described (e.g. quagga). We'd prefer to leave the section in. > "on even on" -> "even on". oops, fixed (r142). Nick _______________________________________________ GROW mailing list [email protected] https://www.ietf.org/mailman/listinfo/grow
