Moin, am 08.02.2019 um 17:24 schrieb Nikolas Pediaditis:
Following your questions, I have some numbers and other information that might be useful.
1. Currently, there will be 977,408 IPv4 addresses remaining in our free pool once we are no longer able to allocate contiguous /22s. This number excludes prefixes that are smaller than a /24 and those prefixes that have been reserved for IXP assignments or temporary assignments. It might also be slightly larger by then, due to addresses that are recovered in the meantime.
2. Over the past three years, we have recovered the following amounts of IPv4 addresses:
2016: 83,712 2017: 106,368 2018: 53,824
Thank you, Nikolas, for the figures. So we're still talking about ~3.8k _new_ LIRs that end up with a /22 worth of addresses in /24s or /23s before 2019-02 would kick in and prolongate the infirmity of IPv4. 3.8k new LIRs that happily can consider starting a business based on IPv4, a legacy technology, and ignore the facts. As Carlos Friaças pointed out on 08.02.2019 at 09:15:
The core purpose of 2019-02 is to allow (more) newcomers to access a tiny bit of IPv4 address space so their (hopefully IPv6-enabled) infrastructure will have path to the IPv4-only world (without going to the market).
Let's put this into perspective: "I see 757979 IPv4 prefixes. This is 238 fewer prefixes than 6 hours ago and 1051 more than a week ago. 57.04% of prefixes are /24. There are 63602 unique originating ASNs. 47266 of these ASNs originate IPv4 only" (Source: https://twitter.com/bgp4_table/status/1097420515206152192) With still about 75% ASNs being IPv4 only, there's definitively no point in prolonging the availability of fresh IPv4 space by reducing the hand-out rate. "IPv4 is over", I hear — so let's be brave and stick to that statement. Regards, -kai