This archive is retained to ensure existing URLs remain functional. It will not contain any emails sent to this mailing list after July 1, 2024. For all messages, including those sent before and after this date, please visit the new location of the archive at https://mailman.ripe.net/archives/list/address-policy-wg@ripe.net/
[address-policy-wg] About the /22 allocation limitation
- Previous message (by thread): [address-policy-wg] About the /22 allocation limitation
- Next message (by thread): [address-policy-wg] About the /22 allocation limitation
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Gert Doering
gert at space.net
Thu Apr 10 13:30:49 CEST 2014
Hi, On Thu, Apr 10, 2014 at 12:02:27PM +0200, Dpto. Datos Television Costa Blanca wrote: > > Doing "more efficient" allocations would have put more load on the routing > > system (because you'd have many more small chunks for those ISPs that have > > grown over time), and you'd *still* run into IPv4 exhaustion. > But instead of running into exhaustion in "2 months" we can handle it to > be "2 years". Please, take in account the time between quotes as an example. True. But will it change anything? We knew that we'd run out of IPv4 at least 10 years ago, and we've made lots of noise to push people towards IPv6 - and it only started for real when IPv4 had run out. So if we had made it "last longer", then the "IPv6 for real" deployment in the large ISPs would have started later - and the installed basis of IPv4-using gear would have been much larger, so the migration would have been *more* work... > Should we take more care in "efficient allocations" or "efficient > routing tables"? > Are in the actually routers with problems in the routings tables in the > same way we have problems with the IPv4 exhaustion? Highly so. Depending on which vendor you used, "typical" gear deployed about 5-7 years ago had a hard limit at 256.000 routes - and that came quite close for a number of ISPs. The hardware upgrade to support 1 million routes did cost significant money (like, 10.000-50.000 EUR/router), and it does not truly support "1 million IPv4" routes, if you also have IPv6 and MPLS in your network - more like 700.000 IPv4 routes in typical deployments. Now, before the big discussion starts: there is other gear in the market that scales up to 2 million, etc., but I wanted to point out that these are real-world hard limits, and the amount of "headroom" we have between "what is out there today" (500k) and "what some of the fairly widely deployed core routers can do today" (700k) is not so big that we want to risk an explosion by factor 2. Gert Doering -- NetMaster -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 811 bytes Desc: not available URL: </ripe/mail/archives/address-policy-wg/attachments/20140410/b94e82c1/attachment.sig>
- Previous message (by thread): [address-policy-wg] About the /22 allocation limitation
- Next message (by thread): [address-policy-wg] About the /22 allocation limitation
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]