Hi all,
I'd like to share a simulation on how prop-113 could possibly affect the consumption of the remaining 103/8 block.
Many thanks to Guangliang and software team from APNIC for providing data on portable assignments registered in APNIC database.
The idea behind this simulation is that by changing the requirement based on needs based through prop-113 (and multihoming not a must),
networks receiving non-portable assignments shorter than /24 today will be eligible to receive direct assignments from APNIC/NIRs.
I took a look at the past data on how many non-portable assignments shorter than /24 were registered.
I calculated how it could affect the consumption of the 103/8, if similar number of assignments were requested to receive direct assignments from 103/8, once they become eligible after prop-113.
Depending on calculation, it looks like prop-113 could accelerate the timeframe of exhaustion of 103/8 from roughly 5.5 years to 3.4 - 2.7 years:
- Today without prop-113, if we continue to allocate roughly 2,000 * /22 blocks per year (2,275 allocations in 2014), the remaining 103 block would be exhausted in 5.5 years.
- If similar no.of requests as non-portable assignments >= /24 as 2014 is made, roughly 3270.75(1270.75 + 2,000) * /22 blocks per year, the block would be exhausted in 3.4 years
- If similar no.of requests as non-portable assignments >= /24 as 2010 is made, roughly 4148.25(2148.25 + 2,000) * /22 blocks per year, the block would be exhausted in 2.7 years
* This count is only based on assignments registered in APNIC and JPNIC database, and does not take into account of assignments registered in other NIRs's database.
* This is assuming that all will be requesting direct assignments from prop-113 and that no. of assignments requests remain flat each year
* Portable assignments made from 103/8 in total is 1,997 blocks (this is not included in the consumption count)
I'd be interested to hear what we think about this possible implications of prop-113.
Do people think expanding the target of direct assignments through prop-113 should be given the higher priority than possible implication accelerating the consumption of 103/8 block?
----
Remaining pool in 103/8:
68% (from APNIC39 update) = approximately 11,408K hosts
No.of /22 allocations from 103/8:
2011 2012 2013 2014
APNIC 856 1098 1542 2275
NON-PORTABLE objects >= /24:
2010: 8593 * /24 (7899 + 694) = 2148.25 * /22
2011: 6295 * /24 (5017 + 1278)
2012: 3678 * /24 (3025 + 653)
2013: 4613 * /24 (3869 + 744)
2014: 5083 * /24 (4052 + 1031) = 1270.75 * /22
* Based on objects registered in APNIC + JPNIC database
* Does not include assignments in other NIRs' database
----
Izumi
On 2015/03/06 10:01, Masato Yamanishi wrote:
Dear colleagues
Version 2 of prop-113: "Modification in the IPv4 eligibility criteria",
did not reach consensus at the APNIC 39 Policy SIG and was returned to
the mailing list for further discussion.
Problem Statement:
Multi-homing is mandatory for small IPv4 delegations to end-sites
(assignments). Requesters might obtain multi-homing when it is not
required, fabricate multi-homing information in their request, or
not apply for the space they need.
Proposal details, including the full text of the proposal, history, and
links to previous versions are available at:
http://www.apnic.net/policy/proposals/prop-113
Regards
Masato
sig-policy mailing list
sig-policy@lists.apnic.net
http://mailman.apnic.net/mailman/listinfo/sig-policy
