Re: [members-discuss] [EXTERNAL] Re: New Charging Scheme
Hi Jørgen, What’s the point of a BCOP if we shouldn’t pay much attention to it? Is the BCOP wrong, or is the allocation policy too rigid? Sadly, the newer generation of stateless IPv6 transition technologies, such as MAP and lw4over6 etc., trade flow tracking complexity for more rigid, complex, and pre-planned addressing plans. My point (and I believe Gert’s also) is that larger IPv6 allocations shouldn’t be difficult when the RIPE NCC is provided with justification. The debate I’ve hijacked this thread for, is what constitutes sufficient justification? -Richard From: members-discuss <members-discuss-bounces@ripe.net> on behalf of Jørgen Hovland <jorgen@ssc.net> Date: Monday, 21 January 2019 at 17:10 Cc: "members-discuss@ripe.net" <members-discuss@ripe.net> Subject: Re: [members-discuss] [EXTERNAL] Re: New Charging Scheme I don't think the point of IPv6 is to spend, spend, spend either. A limited resource will always run out. If you think you are going to need a quadruple billion IP-addresses on every LAN, then sure go a head and spend. I would however not pay much attention to a BCOP that recommends /48 for businesses, /56 for residential, /54 for single moms and /52 for Fortnite players. The moment you start hardcoding prefixlengths into your design, or any number for that matters, you are certainly not going to have a future proof network... Jørgen Information in this email including any attachments may be privileged, confidential and is intended exclusively for the addressee. The views expressed may not be official policy, but the personal views of the originator. If you have received it in error, please notify the sender by return e-mail and delete it from your system. You should not reproduce, distribute, store, retransmit, use or disclose its contents to anyone. Please note we reserve the right to monitor all e-mail communication through our internal and external networks. SKY and the SKY marks are trademarks of Sky Limited and Sky International AG and are used under licence. Sky UK Limited (Registration No. 2906991), Sky-In-Home Service Limited (Registration No. 2067075), Sky Subscribers Services Limited (Registration No. 2340150) and Sky CP Limited (Registration No. 9513259) are direct or indirect subsidiaries of Sky Limited (Registration No. 2247735). All of the companies mentioned in this paragraph are incorporated in England and Wales and share the same registered office at Grant Way, Isleworth, Middlesex TW7 5QD
Hi, On Mon, Jan 21, 2019 at 05:07:24PM +0000, JÃrgen Hovland wrote:
<html><body><div>I don't think the point of IPv6 is to spend, spend, spend either.</div><div>A limited resource will always run out. If you think you are going to need a quadruple billion IP-addresses on every LAN, then sure go a head and spend.
Do the math. A limited resources (IPv6 addresses) distributed over *another* limited resource (humans running networks) will or will not run out, depending on the distribution ratio. As in: if you have 100 cakes and 5 kids, there is no way these kids are going to eat all the cakes, no matter how liberal your cake distribution policies are. We've been distributing IPv6 prefixes for slightly over 20 years now, and we haven't used up 1% of 1/8 of the IPv6 space yet. The growth curve isn't exponential, as the number of "ISP like" entities that will request a prefix from the RIRs is limited - right now, below 20.000 LIRs in the RIPE NCC service region. To give them all a /29 would need half a /14. Now, we have LIRs that are happy with a /32, and we have others that need (or at least "can argue for") a /19 - but those are few, like 1-2 per country. Overall the numbers suggest that all of RIPE land should be fine with a /10 or /9. Which is 1/512 of the total IPv6 space (or 1/64 of FP001 which is all that is available today). Gert Doering -- APWG chair, and math enthusiast -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard, Michael Emmer Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Hi,
Do the math. A limited resources (IPv6 addresses) distributed over *another* limited resource (humans running networks) will or will not run out, depending on the distribution ratio.
As in: if you have 100 cakes and 5 kids, there is no way these kids are going to eat all the cakes, no matter how liberal your cake distribution policies are.
With the start if IPv4 it was thought that this amount of address space would be never run out. Should we do the same error again with IPv6? I am very sure IPv10 or similar will be required in future if we waste address space now (maybe even if we not). No one ever knows what happen in future and the use cases are maybe still out of our possible realization. Just one example: What if in x years there exist robots every part of them communicate with a seperate ip address and one robot could utilize 10.000 addresses or more. I am also a customer of different hosting/colocation Companies and most of them provide me with millions of IPv6 addresses that I not plan to use. 1000 of them would be sufficient for me for very long and if I really need more why just not then give additional nets away. Michael
Hi, On Mon, Jan 21, 2019 at 07:19:05PM +0100, info@cowmedia.de wrote:
Do the math. A limited resources (IPv6 addresses) distributed over *another* limited resource (humans running networks) will or will not run out, depending on the distribution ratio.
As in: if you have 100 cakes and 5 kids, there is no way these kids are going to eat all the cakes, no matter how liberal your cake distribution policies are.
With the start if IPv4 it was thought that this amount of address space would be never run out. Should we do the same error again with IPv6?
Please do the math. Or just look at 100 cakes and try to imagine how many the kids can eat before they really *really* do not want to eat more.
I am very sure IPv10 or similar will be required in future if we waste address space now (maybe even if we not). No one ever knows what happen in future and the use cases are maybe still out of our possible realization. Just one example: What if in x years there exist robots every part of them communicate with a seperate ip address and one robot could utilize 10.000 addresses or more.
I was talking *networks*. Every *network* has an effectively unlimited number of addresses inside, and we have more than enough networks for every human on earth (vastly more than enough).
I am also a customer of different hosting/colocation Companies and most of them provide me with millions of IPv6 addresses that I not plan to use. 1000 of them would be sufficient for me for very long and if I really need more why just not then give additional nets away.
Because we can. And it enables new uses and applications that are unthinkable in a "everything hides between two layer of N:1 NAT" world. If we do not want that, we can just stick to IPv4 - with 100:1 carrier-grade NATs in the access networks, the roughly 3 billion usable IPv4 addresses are sufficient for a consumer-producer-style Internet (they just need to be redistributed better). I don't think this is a desirable path, though. Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard, Michael Emmer Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Hello, I dont full agree. As IPV6 space is not infinite number, the space can over in some moment. Let me give you same example as yours: If you have 5 kids and 100 cakes, it is clear that they cant fully eat the cakes. But if you give all the 100 cakes to the kids, they will bite a little from every cake to taste if there are difference, and in the end you will end up with 0 cakes and lot of uneaten valuable resources. Ivaylo Josifov Varteh LTD Varna Bulgaria On Mon, 21 Jan 2019, Gert Doering wrote:
Hi,
On Mon, Jan 21, 2019 at 05:07:24PM +0000, J?rgen Hovland wrote:
<html><body><div>I don't think the point of IPv6 is to spend, spend, spend either.</div><div>A limited resource will always run out. If you think you are going to need a quadruple billion IP-addresses on every LAN, then sure go a head and spend.
Do the math. A limited resources (IPv6 addresses) distributed over *another* limited resource (humans running networks) will or will not run out, depending on the distribution ratio.
As in: if you have 100 cakes and 5 kids, there is no way these kids are going to eat all the cakes, no matter how liberal your cake distribution policies are.
We've been distributing IPv6 prefixes for slightly over 20 years now, and we haven't used up 1% of 1/8 of the IPv6 space yet. The growth curve isn't exponential, as the number of "ISP like" entities that will request a prefix from the RIRs is limited - right now, below 20.000 LIRs in the RIPE NCC service region. To give them all a /29 would need half a /14. Now, we have LIRs that are happy with a /32, and we have others that need (or at least "can argue for") a /19 - but those are few, like 1-2 per country.
Overall the numbers suggest that all of RIPE land should be fine with a /10 or /9. Which is 1/512 of the total IPv6 space (or 1/64 of FP001 which is all that is available today).
Gert Doering -- APWG chair, and math enthusiast -- have you enabled IPv6 on something today...?
SpaceNet AG Vorstand: Sebastian v. Bomhard, Michael Emmer Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Hi On Mon, Jan 21, 2019 at 08:35:30PM +0200, ivaylo wrote:
I dont full agree. As IPV6 space is not infinite number, the space can over in some moment.
Let me give you same example as yours:
If you have 5 kids and 100 cakes, it is clear that they cant fully eat the cakes. But if you give all the 100 cakes to the kids, they will bite a little from every cake to taste if there are difference, and in the end you will end up with 0 cakes and lot of uneaten valuable resources.
Just do the math, please. All these arguments have been brought forward hundreds and thousands of times in the last 20 years, by people who were just too lazy to run the numbers. I am willing to reconsider my position if we hit 50% utilization of FP001 in my lifetime (*and* buy everyone a beer who comes and says "told you so!"). In this very unlikely event (MATH!), we have burned through half of 1/8 of the overall address space, and have 14 more tries to make a more restrictive policy. (The kids won't try all 100 cakes :-) ) Gert Doering -- NetMaster -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard, Michael Emmer Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
On 21/01/2019 18:07, Jørgen Hovland wrote:
I would however not pay much attention to a BCOP that recommends /48 for businesses, /56 for residential, /54 for single moms and /52 for Fortnite players.
Dear Jørgen, That BCOP document was discussed at length and reached consensus at RIPE IPv6 WG and BCOP TF that includes many operators of various sizes. Can you please tell us a bit more about your concerns? Cheers, Jan
participants (6)
-
Gert Doering -
info＠cowmedia.de -
ivaylo -
Jan Zorz - Go6 -
Jørgen Hovland -
Patterson, Richard (Sky Network Services (SNS))