End Policy for IANA IPv4 allocations to RIRs - AFPUB-2007-v4-002

Ref Name	AFPUB-2007-v4-002	Old Ref.	afpol-v4ep200708
Status	Withdrawn
Date	29 Aug 2007
Author(s)	Akinori MAEMURA, Akira AKAGAWA, Izumi OKUTANI, Kosuke ITO, Kuniaki KONDO, Shuji NAKAMURA, Susumu SATO, Takashi ARANO, Tomohiro FUJISAKI, Tomoya YOSHIDA, Toshiyuki HOSAKA
Policy Affected	The current on-demand global policy for the distribution of IPv4 address space to the Regional Internet Registry (RIR) system.
TOC	* Incentive * Proposal * Rational * Objective * Pros and Cons

Incentive:

This proposal seeks to provide the solutions to the problems in terms of address management which may arise if no measures are taken for IPv4 address exhaustion.

Proposal:

1) Distribute a single /8 to each RIR at the point when new IANA free pool hits 5 */8. This date is defined as "IANA Exhaustion Date".

2) It should be completely left up to each RIR communities to define a regional policy on how to distribute the remaining RIR free pool to LIRs within their respective regions after "IANA Exhaustion Date".

Note 1: It is fine for an RIR to continue operations with the existing policy if that is the consensus decision of the respective RIR community.

Note 2: Address recovery and re-distribution of recovered address space is another important measure for considerations, but should be treated as a separate policy proposal from distribution of new IANA pool.

3) RIRs should provide an official projection on IANA Exhaustion Date to the community through their website, at their Policy Meetings and through any other effective means.

Rationale:

[current problem]

There are two major issues in terms of address management if no measures are taken for IPv4 address exhaustion.

1) Continue applying a global coordinated policy for distribution of the last piece(s) of RIR's unallocated address block does not match the reality of the situation in each RIR region.

Issues each RIR region will face during the exhaustion period vary by region as the level of development of IPv4 and IPv6 are widely different. As a result, applying a global co-ordinated policy may not adequately address issues in a certain region while it could be work for the others.

For example, in a region where late comers desperately need even small blocks of IPv4 addresses to access to the IPv4 Internet, a policy that defines the target of allocations/assignments of IPv4 address space to be late comers would be appropriate in such region. This would allow availablilty of IPv4 address space for such requirements for more years.

Another example comes from difference in IPv6 deployment rate. For a region where IPv6 deployment rate is low, measures may be necessary to prolong IPv4 address life for the existing business as well as for new businesses until networks are IPv6 ready. Some regions may have strong needs to secure IPv4 address space for translators.

A globally coordinated policy which addresses all the issues listed above to meet the needs for all RIR regions may result in not solving issues in any of the regions.

2) LIRs and stakeholders remain unprepared for the situation if they are not informed.

If LIRs and the community are uninformed of the exhaustion, their services and networks remain unprepared to face the situation at the time of exhaustion.

[Objective of the proposal]

This proposal seeks to provide the following solutions to the problems listed above.

1) RIR community should be able to define their own regional policies on how to assign the last piece(s) of allocation block in order to address their own regional issues during the exhaustion period.

2) RIRs should provide official projection of the date when LIRs will be able to receive the allocations under the current criteria. The criteria should remain consistent until this date in order to avoid confusion.

[Pros and Cons]

Pros: + It allows each RIR community to define a policy on how to distribute the last piece(s) of allocations which best matches their situation.

• It helps LIR better informed of the date when they are able to receive allocations from RIRs under the current criteria and prepare for the event.

Cons: + Concerns could be raised about allocating a fixed size to all RIRs, that it artificially fastens the consumption rate of some RIR regions. However, its impact is kept to minimum by keeping the allocation size to a single /8 which makes merely 3-4 months difference.

• Concerns could be raised that explicitly allowing regional policies will encourage RIR shopping. However, this should not happen if the requirements within each region is adequately reflected in each RIR's policy through PDP. RIR may also chose to add criteria to prevent LIRs from other regions submitting such requests.

Timetable for implementation: Immediate after all 5 RIRs (and possibly ICANN) ratifies the policy.