nanog mailing list archives
Re: ULA and RIR cost-recovery
From: Michael.Dillon () radianz com
Date: Thu, 25 Nov 2004 12:46:55 +0000
The problem with this scheme is that it's only aggregatable if there's some POP that lots of carriers connect to in the proper geographic areas. What is the carriers' incentive to show up -- peer? -- at such points, rather than following today's practices?
Leaving aside the specifics of any particular geopgraphic addressing scheme for the moment... If we adopt a geographic addressing scheme for a part of the IPv6 space we are really saying that we expect a part of the IPv6 network topology to be geographically based. While it is convenient to think og geographical divisions in terms of boundaries, in real world networks the geographical divisions are defined by peering points which the real world refers to as "major cities". So if we do adopt a geographic addressing scheme it makes no sense at all for the RIRs to allocate these addresses to entities that happen to be inside a specific geographic boundary. However, it makes perfect sense to allocate these geographic addresses to an entity who is peering at one or more of the peering points within a geographic boundary. This doesn't mean that everybody at the peering point gets geographic addresses but it does mean that organizations who have hierarchical networks with geographically delineated subdivisions can get geographically aggregatable space. For example, let's assume that one of the geographical divisions is the Romance countries. Outside of these countries the geographical address space for this region would consume exactly one routing table slot, no more. Everyone would route the traffic to the nearest network (using non-geographical addresses) which peers at one of the peering points in Paris, Madrid, Lisbon, Milan, Toulouse, etc. The global routing table would be smaller because networks which do not need detailed global visibility will not be using normal PI addresses. Geographic addressing will only work if non-geographic addressing also exists and if the geographic divisions are neither to small nor too large. RIR boundaries are too large. Most national boundaries are too small. With IPv6, routing table size grows 4 times due to the 128 bit addresses. With IPv6 routing table size shrinks because most ASes only need one /32 route. With IPv6 routing table size explodes because most businesses want to multihome. With IPv6 and geographical addressing, routing table size shrinks because most multihomed businesses only need global visibility of their routes within a larger geographical aggregate. --Michael Dillon
Current thread:
- Re: who gets a /32 [Re: IPV6 renumbering painless?], (continued)
- Re: who gets a /32 [Re: IPV6 renumbering painless?] Owen DeLong (Nov 19)
- Re: who gets a /32 [Re: IPV6 renumbering painless?] Adi Linden (Nov 19)
- Re: who gets a /32 [Re: IPV6 renumbering painless?] Stephen Sprunk (Nov 19)
- Re: who gets a /32 [Re: IPV6 renumbering painless?] Owen DeLong (Nov 19)
- ULA and RIR cost-recovery John Curran (Nov 22)
- RE: ULA and RIR cost-recovery Tony Hain (Nov 23)
- RE: ULA and RIR cost-recovery Owen DeLong (Nov 23)
- RE: ULA and RIR cost-recovery Tony Hain (Nov 24)
- Re: ULA and RIR cost-recovery Steven M. Bellovin (Nov 24)
- RE: ULA and RIR cost-recovery Tony Hain (Nov 24)
- Re: ULA and RIR cost-recovery Michael . Dillon (Nov 25)
- Re: geography to get PI in v6 (was: ULA and RIR cost-recovery) Iljitsch van Beijnum (Nov 25)
- RE: ULA and RIR cost-recovery Måns Nilsson (Nov 29)
- Re: ULA and RIR cost-recovery Daniel Roesen (Nov 29)
- Re: ULA and RIR cost-recovery Leo Bicknell (Nov 29)
- Re: ULA and RIR cost-recovery Owen DeLong (Nov 29)
- Re: ULA and RIR cost-recovery Leo Bicknell (Nov 29)
- Re: ULA and RIR cost-recovery Pekka Savola (Nov 29)
- Re: ULA and RIR cost-recovery Owen DeLong (Nov 29)
- Re: ULA and RIR cost-recovery Pekka Savola (Nov 29)
- Re: ULA and RIR cost-recovery Owen DeLong (Nov 30)