Measuring Roaming in Europe: Infrastructure and Implications on Users’ QoE

by |Published

This repository contains data and information regarding the paper:

Anna Maria Mandalari, Andra Lutu, Ana Custura, Ali Safari Khatouni, Özgü Alay, Marcelo Bagnulo,
Vaibhav Bajpai, Anna Brunstrom, Jörg Ott, Martino Trevisan, Marco Mellia, Gorry Fairhurst. “Measuring Roaming in Europe: Infrastructure and Implications on Users’ QoE“. Currently under revision

Summary

“Roam like Home” is the initiative of the European Commission (EC) to end the levy of extra charges when roaming within the European region. As a result, people can use data services more freely across Europe. However, the implications of roaming solutions on network performance have not been carefully examined. This paper provides an in-depth characterization of the implications of international data roaming within Europe. We build a unique roaming measurements platform using 16 different mobile networks deployed in 6 countries across Europe. Using this platform, we measure different aspects of international roaming in 4G networks in Europe, including mobile network configuration, performance characteristics, and quality of experience. We find that operators adopt a common approach to implement roaming called Home-routed roaming (HR). This results in additional latency penalties of 60 ms or more, depending on geographical distance. This leads to worse browsing performance, with a significant increase in the metrics related to Quality of Experience (QoE) of users, namely Page Load time and Speed Index. We further analyze in isolation the impact of latency on QoE metrics and show that the penalty imposed by HR leads to degradation on QoE-metrics in the order of 20-150% depending on the geographical distance.

Different topology architectures that can be used for roaming in a mobile network, namely, HR, local breakout (LBO) and IPX hub breakout (IHBO).

Implications of roaming

The HR data implies that the roaming user’s exit point to the Internet is always in the original home country. Depending on the location of the server, this translates to a potential delay and performance penalty. The largest delay penalty occurs when the roaming user tries to access a server located in the visited country. This is because the packets must go back and forth from the home network. Surprisingly, we note that the HR configuration also impacts the case when the roaming user accesses a target server located in the home network.

ECDF of the RTT from mobile nodes to target servers

There is a sizable penalty that roaming users suffer, which varies depending on the country where they travel. It is particularly interesting as QoE metrics degradation is known to impair not only users’ QoE, but also the business model of big Internet players. Indeed, even small deterioration of quality levels could result in losses of revenues to providers. For all websites, the performance is lower in the order of 15-20%.

Empirical distribution ot the page load time with and without roaming for different target websites.

The dataset

The dataset is available at https://mplanestore.polito.it:5001/sharing/DAtRp5VNk.

You find different CSV files, one for each measurement campaign. In particular, you find:

  • Latency measurements (traceroute.csv): periodic traceroute measurements against all the servers we deploy in each country as measurement responders.
  • DNS resolutions (dns.csv): DNS lookups (over UDP port 53) against a list of 180 target Fully Qualified Domain Names (FQDNs) mapped to advertisement services.
  • Web Browsing (browsing.csv): automatic visits to 100 target pages using the browsertime dockerized automatic browser. We collected a large number related to page load time.
  • Emulated Web Browsing (browsing_emulated.csv): We repeated the visits as above on cabled servers, while varying artificially the latency to test a larger number of delay curves, without the scalability limitations of mobile nodes.