Fw: [ppml] Re: ARIN Policy Proposal 2002-7

[Charles Scott]

On Tue, 8 Oct 2002, Newton, Justin wrote:

> You could do that.  In this case you are still restricted to outages caused
> by your routing provider.  I.e. if the aggregator goes down, you are off the
> network.  This technique would isolate you from physical level problems, but
> would not help you with logic or code problems in the routing
> infrastructure.  You would also then, instead of being tied to an ISP, be
> tied to an aggregator, which gets us right back where we started.

  You are always at risk from routing errors. So I guess it all depends on
who is the aggrigator. If the aggrigator is reliable and well placed at a
variety of NAP's, then it should be pretty much as reliable as any
conventional multi-homing. I would even go so far as to say that ARIN
could consider building-out this service as a means to solve the route
table issues and still offer long prefix IP blocks, or, it could be a
service provided by some new association or mutually non-offensive
organization (there may even be some revenue stream). It just seems that
if there's going to be some problems with long prefix allocations, that it
would be best to concentrate concern with those problems on the smaller
set of entities. In otherwords, wouldn't it be better to deal with the
technical concerns of the aggrigator and it's customers than have everyone
on the network deal with a significant expansion of routing tables.

Chuck Scott
cscott at gaslightmedia.com