Characterizing residential broadband networks

Abstract

A large and rapidly growing proportion of users connect to the Internet via residential broadband networks such as Digital Subscriber Lines (DSL) and cable. Residential networks are often the bottleneck in the last mile of today's Internet. Their characteristics critically affect Internet applications, including voice-over-IP, online games, and peer-to-peer content sharing/delivery systems. However, to date, few studies have investigated commercial broadband deployments, and rigorous measurement data that characterize these networks at scale are lacking.In this paper, we present the first large-scale measurement study of major cable and DSL providers in North America and Europe. We describe and evaluate the measurement tools we developed for this purpose. Our study characterizes several properties of broadband networks, including link capacities, packet round-trip times and jitter, packet loss rates, queue lengths, and queue drop policies. Our analysis reveals important ways in which residential networks differ from how the Internet is conventionally thought to operate. We also discuss the implications of our findings for many emerging protocols and systems, including delay-based congestion control (e.g., PCP) and network coordinate systems (e.g., Vivaldi).

Characterizing Home Network Traffic: An Inside View

“…As residential broadband users continue to grow, many studies have been devoted to measure and characterize residential broadband networks [7][8][9]. However, all of these studies stand from outside home networks, and lack the visibility of the home networks, such as home network architecture, diversity of end hosts.…”

“…For example, [7] examines the growth of residential user-to-user traffic in Japan, a country with a high penetration rate of residential broadband access, and studies the impact of these traffic on usage patterns and traffic engineering of commercial backbone networks. In addition, [8] studies several properties of broadband networks, including link capacities, round-trip times, jitter, and packet loss rates using active TCP and ICMP probes, while [9] passively collects packet-level traffic data of residential networks at aggregated routers of a large Internet service provider, and analyzes dominant characteristics of residential traffic including network and transport-level features, prominent applications, and network path dynamics. Different from these prior work, this paper leverages the availability of traffic flows exported from programmable home routers, and presents the first study of traffic characteristics of Internet-capable devices in home networks.…”

“…Several recent studies have been devoted to understanding traffic characteristics of home networks using aggregated and sampled traffic collected from edge routers in Internet service providers [7][8][9]. However, these measurement studies stand from the perspective of outside home networks, thus lack the visibility of what is happening in home networks.…”

BSense: A Flexible and Open-Source Broadband Mapping Framework

“…Flexibility is also important to accommodate the wide range of broadband access technologies and to factor in latest advances and best practice in terms of measurement techniques. Moreover, these efforts take one of two different approaches we refer to as model based (e.g., [7][8][9]) and measurement based (e.g., [10,11,5,12,6,13,14]) even though neither is enough -model based approaches fail to capture the discrepancy between expected and actual broadband quality experienced by consumers, whereas measurement based approaches are clearly not useful for identifying notspots. We further discuss these different broadband mapping approaches in §2.…”

“…Broadband quality assessment in BSense is based on hostbased continuous measurement stream obtained using our multi-platform and technology-adaptive software client, which periodically runs as a background process on users' computers. As elaborated in §2, the measurement paradigm employed in BSense results in a lower-cost and flexible alternative to [10,11], and reduces measurement bias compared to [5,12,14]. -Unique to BSense is a flexible broadband quality index for summarizing the collective effect of various underlying attributes such as download/upload speeds and latency.…”