Browse Prior Art Database

The H-Density Ratio for Address Assignment Efficiency An Update on the H ratio (RFC3194) Disclosure Number: IPCOM000005941D
Original Publication Date: 2001-Nov-01
Included in the Prior Art Database: 2001-Nov-21
Document File: 8 page(s) / 15K

Publishing Venue

Internet Society Requests For Comment (RFCs)

Related People

A. Durand: AUTHOR [+2]


This document provides an update on the "H ratio" defined in RFC 1715. It defines a new ratio which the authors claim is easier to understand.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 24% of the total text.

Network Working Group                                          A. Durand

Request for Comments: 3194                              SUN Microsystems

Updates: 1715                                                 C. Huitema

Category: Informational                                        Microsoft

                                                           November 2001

       The Host-Density Ratio for Address Assignment Efficiency:

                        An update on the H ratio

Status of this Memo

   This memo provides information for the Internet community.  It does

   not specify an Internet standard of any kind.  Distribution of this

   memo is unlimited.

Copyright Notice

   Copyright (C) The Internet Society (2001).  All Rights Reserved.


   This document provides an update on the "H ratio" defined in RFC

   1715.  It defines a new ratio which the authors claim is easier to


1. Evaluating the efficiency of address allocation

   A naive observer might assume that the number of addressable objects

   in an addressing plan is a linear function of the size of the

   address.  If this were true, a telephone numbering plan based on 10

   digits would be able to number 10 billion telephones, and the IPv4 32

   bit addresses would be adequate for numbering 4 billion computers

   (using the American English definition of a billion, i.e. one

   thousand millions.) We all know that this is not correct: the 10

   digit plan is stressed today, and it handles only a few hundred

   million telephones in North America; the Internet registries have

   started to implement increasingly restrictive allocation policies

   when there were only a few tens of million computers on the Internet.

   Addressing plans are typically organized as a hierarchy: in

   telephony, the first digits will designate a region, the next digits

   will designate an exchange, and the last digits will designate a

   subscriber within this exchange; in computer networks, the most

   significant bits will designate an address range allocated to a

   network provider, the next bits will designate the network of an

   organization served by that provider, and then the subnet to which

   the individual computers are connected.  At each level of the

Durand & Huitema             Informational                      [Page 1]

RFC 3194                An update on the H ratio           November 2001

   hierarchy, one has to provide some margins:  one has to allocate more

   digits to the region code than the current number of regions would

   necessitate, and more bits in a subnet than strictly required by the

   number of computers.  The number of elements in any given level of

   the   hierarchy will change over time, due to growth and mobility.

   If the current allocation is exceeded, one has to engage in

   renumbering, which is painful and expensive.  In short, trying to

   squeeze too many objects into a hierarchical address space increases

   the level of pain endured by operators and subscribers.

   Back in 1993, when we were debating the revision of the Internet

   Protocol, we wondered what the acceptable ratio of utilization was of

   a given addressing plan.  Coming out with such a ratio was useful to

   assess how many computers could be connected to the Internet with the

   current 32-bit addresses, as well as to decide the size of the next

   generation addresses.  The second point is now deci...