[arin-ppml] IPv6 Allocation Planning

[Owen DeLong]

On Aug 9, 2010, at 7:27 PM, William Herrin wrote:

> On Mon, Aug 9, 2010 at 8:23 PM, Joel Jaeggli <joelja at bogus.com> wrote:
>> There are quite a few people on this list who've had deployed dual stack
>> ipv6 networks for longer than it took us to put a man on the moon.
>> 
>> Jaded rather than inexperienced is the first thing that some to mind.
> 
> Hi Joel,
> 
> There were plenty of jaded IPv4 gurus in 1993. Those that survived and
> were still acknowledged gurus were far wiser and more knowledgeable in
> 2003.
> 
> My point is: there's a lot about IPv6 policy we can't yet predict with
> a high confidence. We'll do the best we can and we'll change it when
> we learn better, but for now there's still a lot of guesswork
> involved.
> 

ARIN IPv6 policy mostly dates back to before 2004 when the NRPM was
first introduced. Much of it, I suspect, goes back nearly to the beginning
of ARIN over 10 years ago.

The following are the changes since 2004:

2003-4 8/3/04 section 6.5.1.1.d modified 	Expand ISP planning horizon to 5 years (from 2)
										Clarify qualifying LIRs.

2005-5 2/12/06 sections 6.5.2.2 and 6.7 modified
										Revise HD ratio requirement from .8 to .94

2005-1 5/9/06 sections 6.5.4 modified, 6.5.8 created
										Introduction of PI end-user assignments for IPv6

2005-8 5/9/06 Sections 6.1.1, 6.2.7, 6.5.1-5, 6.7, 6.8.3, and 6.9 modified
										End site from /48 in policy to LIR or ISP internal decision
										
-- Editorial update to 6.6 on 7/27/06

2006-2 1/6/07 Section 6.10 created			Microallocations for internal infrastructure

2007-4 6/14/07 Section 6.1.1 modified		Removal of the term "INTERIM" from IPv6 policy

2007-25 12/11/07 Sections 6.5.1.1.d, 6.5.3, 6.5.4.4, 6.5.8.2, and 6.5.8.3 modified
										Proposal titled literally "IPv6 Policy housekeeping".
										Mostly editorial changes and no real policy modifications
										Entirely stuff that would be handled today by staff with a
										consent decree from the Advisory Council.

2007-21 7/8/08 Section 6.5.8.1 modified	IPv6 for Legacy Holders with RSA and Efficient Use

-- 9/23/09 Editorial updates to sections 6 and 6.2

2008-3 12/18/09 Section 6.5.9 created		Community Networks IPv6 Assignment

2009-5 12/18/09 Section 6.11 created		IPv6 Multiple Discreet Networks

2009-7 12/18/09 Section 6.5.1.1 modified	Remove routing advice from IPv6 policy

--1/13/10 Editorial updates to 6.1.1, 6.2, 6.5.4, 6.5.5, 6.6, and 6.8


So, with the exception of those 11 policy changes, our IPv6 policy is now
probably close to 10 years old. I'd say we should be applying what we
have learned in the interim to policy today with an eye towards applying
more knowledge to policy in the more distant future.

Certainly, I think we can do better than current IPv6 policy with the
knowledge and experience gained over the last 10 years.

Some lessons include:

	Humans are not good at math.
	Humans are really not good at bit-math.
	Putting things on nibble boundaries reduces human mistakes.
	Giving out large prefixes early allows greater aggregation than giving out
		small prefixes often.
	Giving out large prefixes early will not create an addressing shortage for
		any foreseeable (and many unforeseeable) definitions of later.

Owen