MAPPING -- Measuring And Practically Predicting INternet Growth

Measuring and predicting growth in Internet addressing, routing complexity and energy usage

Overview

The MAPPING research program was initially named STING (Surveying The INternet's Growth). We renamed the program in April 2014.

The MAPPING program began in 2009 with a research grant from APNIC Pty Ltd. Some of our initial work focused on address utilisation patterns discernible from game server discovery traffic. Since the middle of 2011 MAPPING has became the home of a new ARC Linkage Project, jointly supported by APNIC Pty Ltd and the Australian Research Council (ARC). The current focus of the program is to develop a better understanding of the migration from IP version 4 (IPv4) to its successor IP version 6 (IPv6) and the migration's potential impacts on factors, such as the cost of Internet access and the size of "routing tables".

Nearly all IPv4 address prefixes have been allocated, yet the migration to IPv6 is slow. A reason for this could be that not all allocated IPv4 addresses are actually used, but how much of the allocated address space is still unused remains unclear. The potential consequences over the next few years are an increasing uptake of IPv6, increased Network Address Translation (NAT) deployment, and a new marketplace for trading blocks of unused IPv4 address space. However, how exactly the IPv4 to IPv6 migration will unfold is still poorly understood.

Estimating the rate of consuming IPv4 addresses, the proportion of allocated but underutilised IPv4 address space and the actual number of hosts (including hosts behind NATs), will allow predicting the likely value and costs of an international IPv4 address market (potentially increasing the cost of Internet service), developing strategies for distribution of remaining IPv4 addresses, and setting the time frame of IPv6 deployment. Furthermore, estimating changes in fragmentation of address use, will allow identifying the potential for overflow of "routing tables" in core infrastructure.

Another facet of this program is the growing demand for Internet-related electrical energy worldwide -- both in core infrastructure and end user devices. We aim to improve current estimates of end-user consumption that are based on unwieldy manual surveys.

Project Goals

• Develop innovative active probing and passive monitoring techniques to study and predict daily and multi-year changes in current use of the IPv4 address space
• Develop plausible models for the current use and future demand of Internet addresses
• Track IPv4 address utilisation to improve society's ability to learn lessons applicable to IPv6 roll-out policies and guide regulators in understanding the future market for IPv4 addresses
• Estimate the actual number of hosts by estimating how many addresses are NATed addresses and how many hosts are behind NATs
• Improve previous energy consumption estimates by providing a tighter bound on the number of devices active at any given time

As part of this project we will develop and release tools to assist in data gathering and analysis, and publish interim results and papers on our website. The links at the top will take you to additional information.

Outcomes

From 2009 to 2014 we have:

• Developed a technique to infer the time zones of prefixes and autonomous systems by monitoring game server discovery traffic [paper]
• Visualised the global distribution of game clients (see animations here)
• Developed novel techniques to measure the IPv6 capabilities of Internet clients and mitigate the measurement sampling error (see our IMC 2012 paper here)
• Created an IPv4 Internet Map to visualise the used IPv4 address space (Update: there is now a time series of IPv4 maps)
• Created an IPv6-capable IPv4 Internet Map to visualise the IPv4 address space that is IPv6-capable
• Developed a technique based on Capture-recapture (CR) to estimated the actively used IPv4 space and computed estimates over several years based on several collected IPv4 datasets (see our IMC 2014 paper here and also see here)
• Developed a secure CR technique that allows estimating the used IPv4 space from IPv4 address data of multiple collaborators while not revealing any addresses observed by one collaborator to the other parties; a published a proof-of-concept implementation can be downloaded from tools

Are you behind a NAT? Find out here.

Chief Investigators
Grenville Armitage
Lachlan Andrew
Geoff Huston

Project Members
George Michaelson
Sebastian Zander

Alumni
Mattia Rossi

Funding

This project has been made possible in part by grants from APNIC for a project titled "Exploring the Utilisation of IPv4 Address Space and Size of the NATed IPv4 Internet" and an ARC linkage grant with APNIC as partner organisation for a project titled "Tools and models for measuring and predicting growth in internet addressing and routing complexity" (project LP110100240). The research has also been supported by Australian Research Council grant FT0991594.

Collaboration

This project acknowledges the support of APNIC Pty Ltd and Valve Corporation in helping provide insights into the utilisation of Internet address space as seen by online services.