Re: I-D ACTION:draft-huston-hd-metric-01.txt

[Tim Chown <tjc@ecs.soton.ac.uk>]

Hi Fred,

These sem to be very operational issues to me.  I think both are views
of the same overall issue, i.e. evolving the 'big picture' for BCP in
IPv6 address allocation and management.   Thomas cites the HD ratio
in his draft, for example, but not Geoff's draft (I think!).

I think an interesting bit of RFC3177 is the reasoning for a fixed /48
boundary in section 3.  Issues like multihoming and site-locals that
are discussed, but I don't think Thomas picks those up.  3177 talks of 
leaving the option to retrofit an 8+8 multihoming model, but I think 
we're now heading a different way with shim6.  3177 also talks about 
site locals, which would lead to the implication that we'll need to
modify ULAs to be /56 friendly if Thomas' suggestions are adopted.

Tim

On Thu, Sep 01, 2005 at 11:21:05AM -0700, Fred Baker wrote:
> Are you of the opinion that we should take this, and perhaps Thomas'  
> draft, up as a v6ops draft?
> 
> (I would question whether free advice on address allocation policy is  
> actually an IPv6 WG topic as much as an operational topic anyway, and  
> certainly comments on the HD ratio is an operational topic)
> 
> On Sep 1, 2005, at 8:39 AM, Tim Chown wrote:
> >On Wed, Aug 31, 2005 at 03:50:01PM -0400, Internet-Drafts@ietf.org  
> >wrote:
> >
> >>A New Internet-Draft is available from the on-line Internet-Drafts  
> >>directories.
> >>    Title        : Considerations on the IPv6 Host density Metric
> >>    Author(s)    : G. Huston
> >>    Filename    : draft-huston-hd-metric-01.txt
> >>    Pages        : 18
> >>    Date        : 2005-8-31
> >>
> >>This memo provides an analysis of the Host Density metric as
> >>   currently used to guide registry allocations of IPv6 unicast  
> >>address
> >>   blocks.  This document contrasts the address efficiency as  
> >>currently
> >>   adopted in the allocation of IPv4 network addresses and that  
> >>used by
> >>   the IPv6 protocol.  It is noted that for large allocations  
> >>there are
> >>   very significant variations in the target efficiency metric  
> >>between
> >>   the two approaches.
> >>
> >>A URL for this Internet-Draft is:
> >>http://www.ietf.org/internet-drafts/draft-huston-hd-metric-01.txt
> >>
> >
> >A couple of observations.
> >
> >I note from RFC3194 that it says "The examples suggest an HD-ratio  
> >value
> >on the order of 85% and above correspond to a high pain level, at  
> >which
> >operators are ready to make drastic decisions" and that "...this  
> >suggests
> >that values of 80% or less corresponds to comfortable trade-offs  
> >between
> >pain and efficiency."
> >
> >So the argument here is that very large networks don't share the same
> >HD ratio property?  I think it would be nice to state the crux of  
> >the 'case'
> >of this draft in the intro section.
> >
> >I guess the references for 3513 and 3177 should point to the -bis  
> >versions
> >currently in draft?
> >        draft-ietf-ipv6-addr-arch-v4-04
> >        draft-narten-ipv6-3177bis-48boundary-00
> >
> >This draft states assumptions about /48's, so should probably  
> >discuss the
> >impact of /56's being the default?   Or do you think Thomas' draft  
> >should
> >discuss this?   The two seem quite linked :)
> >
> >-- 
> >Tim/::1
> >
> >

-- 
Tim/::1