[Community-Discuss] IPv6 Chapter 254

[Andrew Alston]

Heh,

We all have our worries about Cloud Innovation – but if you’re going to give stats to the list – at least make them accurate.

Cloud innovation according to the AfriNIC database has 2 /12s and 1 /11, which is 33% below what you claim in your email below.

Some interesting things about this space though.  The majority of it (shockingly) is announced out of South Africa and visible via the South African internet exchanges.  *Shrug*

If they could meet the policy requirements to get the space – and we the community don’t like the fact that they have it – we have only one group of people to blame, ourselves, because we didn’t change the policies.  The community has no place to complain if someone meets the policy that they did not change.

I still stand by my opposition of the soft-landing-biz policy, if we want to make changes to the actual ip allocation policies to ensure space is used adequately, great, but I oppose anything that stops organizations that legitimately need the space getting whatever space they can *legitimately use on the continent*

And quite frankly – if the AfriNIC pool depletes – we’re all better off because we’ll get off our asses and do what we should be doing – deploying IPv6.

It is sickening that there is *ONE* country on the continent with a V6 penetration level higher than 0.6% (Zimbabwe, 4.92% as of latest stats), two that are higher than 0.4% (Sudan is after Zimbabwe, at 0.59%), and only 7 other countries on the entire continent in the top 100 in terms of penetration rates worldwide (Botswana (0.38%),  Egypt (0.26%), South Africa (0.23%), Tanzania (0.10%), Cameroon (0.06%), Rwanda (0.05%) and Kenya (0.02%)).  Africa’s V6 penetration levels as a combined total sit at 0.15%.

And the argument about lack of infrastructure and lack of normal internet penetration holds negative water – because the lower the amount of infrastructure and internet penetration in the country the easier it is to swing the pendulum for v6 as percentage of penetration.

While we sit and watch *massive* providers (T-Mobile and Rogers to name two) going *v6 only* other than for translation purposes to v4 – which will cause huge problems for people who have no v6 for various reasons, we sit and worry about depleting v4?  While the global stats CLEARLY demonstrate that v4 depletion = accelerated v6 deployment (using APNIC stats: Americas = 17.19%, Europe = 11.87%, Oceania = 6.49%, Asia = 4.02%, ALL have run out, Africa which still has v4 = 0.16%)

And lets look at those stats above, Americas, the majority is in North America with no soft landing and a hard run out = 17.19%, Europe, which had a soft landing and is still handing out /22 under it = 11.87%, Oceania/Asia (both under APNIC rules) at 6.49% and 4.02% have limited allocations because of restrictions in place since 15th April 2011 and additional space since 27th May 2014.

What does this indicate – no v4 space = v6 deployment.  And yeah, I know, we can argue that well, soft landing has helped spur it in Europe and Asia to a degree – that’s true, and soft landing in Africa may help us spur it – but spur it is not what we need right now – we are already running at a penetration level of 0.16% as against a global average of 12% - what we need is a flat out end of space and people to wake up and go and do v6 instead of talking about it.

I’ve also heard the argument that the bigger providers can worry about v6 and the smaller ones can have the v4 – that’s been stated at meetings many times.  Well, guess what, speaking from a bigger provider perspective, we’ve *DONE* that, and it’s actually far more difficult for a large provider to do that than it is for a small provider because the amount of infrastructure change and network modification to retro-fit v6 to a large network is well… larger.

So, I want to issue a challenge for every single person who is supporting tightening the rules on IPv4 and putting the soft landing policy in place.  What are the ASN’s *YOU* are running IPv6 on, put them on the list, and let us see you leading by example in V6 rollouts before attempting to change the rules for v4 for the rest of the continent.

I’ll start, our primary ASN with v6 behind it is AS30969.  It has over 10 thousand /48s in the whois database, and according to APNIC statistics pushes in excess of 20% v6 penetration levels.

Who’s going to stand up next?

Andrew

From: Omo Oaiya <Omo.Oaiya at wacren.net>
Date: Friday, 14 October 2016 at 18:39
To: Noah <noah at neo.co.tz>
Cc: General Discussions of AFRINIC <community-discuss at afrinic.net>
Subject: Re: [Community-Discuss] IPv6 Chapter 254


On 14 October 2016 at 12:39, Noah <noah at neo.co.tz<mailto:noah at neo.co.tz>> wrote:

> 4)      We know that v4 exhaustion as per other regions forces v6 adoption
>

True story.

>
> Basically a win-win situation. Let the v4 space go to LIRs who want/need it.
>

No, we need to be careful here especially when we hit softlanding.

There is a looming IPv$ tranfer market and fake cloud companies out there.

Noah

I don't understand the hoohah about depletion.   Even apparently real cloud companies are depleting what's left pretty fast.

According to September stats, Cloud Innovation just got another /11 so now has a total of 3 /11 with zero IPv6 and no ASN.  Go figure!

-Omo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/community-discuss/attachments/20161014/0c84fa2c/attachment.html>