[arin-ppml] Policy for IANA to hand out v4 post exhaustion question.

Hannigan, Martin marty at akamai.com
Fri May 21 10:44:37 EDT 2010



From: Raul Echeberria <raul at lacnic.net>
Date: Thu, 20 May 2010 20:02:25 -0300
To: Martin Hannigan <marty at akamai.com>
Cc: Bill Darte <BillD at cait.wustl.edu>, Ted Mittelstaedt <tedm at ipinc.net>,
<arin-ppml at arin.net>
Subject: Re: [arin-ppml] Policy for IANA to hand out v4 post exhaustion
question.



Marty, 

>> Marty's text calls for a equally sized allocation for each RIR. I'm
>> interested in knowing if this meets some 'fairness' test and is
>> congruence with needs-based allocation policy.
> 
> 
>> One thought was linking the IANA allocations unit to the smallest policy
>> based allocation unit of all RIRs. The needs/eligibility based portion of
>> this might create a requirement such as an RIR needs to be "exhausted"
>> before they can become eligible to participate in a round of allocations
>> from the IANA. 

>In fact, this is my interpretation of what is said in the original text of >the
global policy proposal

>"A RIR is eligible to receive additional IPv4 address space from the IANA
>when the total of its IPv4 address holdings is less than 50% of the
>current IPv4 allocation unit, and providing that it has not already
>received an IPv4 allocation from the IANA during the current IPv4
>allocation period."

>I have said that many times during the discussion. Those RIRs that have IP
>addresses for providing services in their regions for more time, will not be
>eligible for receiving allocations from this new pool for a long time.

>It is not exactly the same that you say, since the requirement is not that >the
RIR has to be "exhausted", but "almost exhausted".


50% of a /8 is quite different than 50% of a /20. That is still somewhat
lopsided since a /20 in the LACNIC region "may" last longer than a /20 in
the ARIN region, but at some point this will converge and be equal in all
regions. The minor differential in the beginning is not as remarkable as 50%
of a /8 for example. Near fully exhausting also adds predictability.


> So if an RIR was allocated a /20 based on the min alloc unit
> standard they would have to exhaust that allocation before being eligible
> for another allocation. That could be fair since any RIR not exhausting
> would be skipped over allowing other needs to be fairly fufilled.


>While I think that it is the spirit of the proposal, i would not object being
>more restrictive as you propose.


Excellent. Thank you.

Best,

-M<




More information about the ARIN-PPML mailing list