WiFi, LTE, or both? Measuring multi-homed wireless internet performance

Shuo Deng, Ravi Netravali, Anirudh Sivaraman, Hari Balakrishnan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Over the past two or three years, wireless cellular networks have become faster than before, most notably due to the deployment of LTE, HSPA+, and other similar networks. LTE throughputs can reach many megabits per second and can even rival WiFi throughputs in some locations. This paper addresses a fundamental question confronting transport and application-layer protocol designers: Which network should an application use? WiFi, LTE, or Multi-Path TCP (MPTCP) running over both? We compare LTE and WiFi for transfers of different sizes along both directions (i.e. the uplink and the downlink) using a crowdsourced mobile application run by 750 users over 180 days in 16 different countries. We find that LTE outperforms WiFi 40% of the time, which is a higher fraction than one might expect at first sight. We measure flow-level MPTCP performance and compare it with the performance of TCP running over exclusively WiFi or LTE in 20 different locations across 7 cities in the United States. For short flows, we find that MPTCP performs worse than regular TCP running over the faster link; further, selecting the correct network for the primary subflow in MPTCP is critical in achieving good performance. For long flows, however, selecting the proper MPTCP congestion control algorithm is equally important. To complement our flow-level analysis, we analyze the traffic patterns of several mobile apps, finding that apps can be categorized as "short-flow dominated" or "long-flow dominated". We then record and replay these patterns over emulatedWiFi and LTE links. We find that application performance has a similar dependence on the choice of networks as flow-level performance: An application dominated by short flows sees little gain from MPTCP, while an application with longer flows can benefit much more from MPTCP - if the application can pick the right network for the primary subflow and the right choice of MPTCP congestion control.

Original languageEnglish (US)
Title of host publicationIMC 2014 - Proceedings of the 2014 ACM
PublisherAssociation for Computing Machinery
Pages181-194
Number of pages14
ISBN (Electronic)9781450332132
DOIs
StatePublished - Nov 5 2014
Event2014 ACM Internet Measurement Conference, IMC 2014 - Vancouver, Canada
Duration: Nov 5 2014Nov 7 2014

Other

Other2014 ACM Internet Measurement Conference, IMC 2014
CountryCanada
CityVancouver
Period11/5/1411/7/14

Fingerprint

Internet
Application programs
Throughput
Network protocols

Keywords

  • LTE
  • Mobile device
  • Multi-network
  • Multi-path TCP

ASJC Scopus subject areas

  • Software
  • Computer Networks and Communications

Cite this

Deng, S., Netravali, R., Sivaraman, A., & Balakrishnan, H. (2014). WiFi, LTE, or both? Measuring multi-homed wireless internet performance. In IMC 2014 - Proceedings of the 2014 ACM (pp. 181-194). Association for Computing Machinery. https://doi.org/10.1145/2663716.2663727

WiFi, LTE, or both? Measuring multi-homed wireless internet performance. / Deng, Shuo; Netravali, Ravi; Sivaraman, Anirudh; Balakrishnan, Hari.

IMC 2014 - Proceedings of the 2014 ACM. Association for Computing Machinery, 2014. p. 181-194.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Deng, S, Netravali, R, Sivaraman, A & Balakrishnan, H 2014, WiFi, LTE, or both? Measuring multi-homed wireless internet performance. in IMC 2014 - Proceedings of the 2014 ACM. Association for Computing Machinery, pp. 181-194, 2014 ACM Internet Measurement Conference, IMC 2014, Vancouver, Canada, 11/5/14. https://doi.org/10.1145/2663716.2663727
Deng S, Netravali R, Sivaraman A, Balakrishnan H. WiFi, LTE, or both? Measuring multi-homed wireless internet performance. In IMC 2014 - Proceedings of the 2014 ACM. Association for Computing Machinery. 2014. p. 181-194 https://doi.org/10.1145/2663716.2663727
Deng, Shuo ; Netravali, Ravi ; Sivaraman, Anirudh ; Balakrishnan, Hari. / WiFi, LTE, or both? Measuring multi-homed wireless internet performance. IMC 2014 - Proceedings of the 2014 ACM. Association for Computing Machinery, 2014. pp. 181-194
@inproceedings{3f43a922dded4554a13eca443aacc3c1,
title = "WiFi, LTE, or both? Measuring multi-homed wireless internet performance",
abstract = "Over the past two or three years, wireless cellular networks have become faster than before, most notably due to the deployment of LTE, HSPA+, and other similar networks. LTE throughputs can reach many megabits per second and can even rival WiFi throughputs in some locations. This paper addresses a fundamental question confronting transport and application-layer protocol designers: Which network should an application use? WiFi, LTE, or Multi-Path TCP (MPTCP) running over both? We compare LTE and WiFi for transfers of different sizes along both directions (i.e. the uplink and the downlink) using a crowdsourced mobile application run by 750 users over 180 days in 16 different countries. We find that LTE outperforms WiFi 40{\%} of the time, which is a higher fraction than one might expect at first sight. We measure flow-level MPTCP performance and compare it with the performance of TCP running over exclusively WiFi or LTE in 20 different locations across 7 cities in the United States. For short flows, we find that MPTCP performs worse than regular TCP running over the faster link; further, selecting the correct network for the primary subflow in MPTCP is critical in achieving good performance. For long flows, however, selecting the proper MPTCP congestion control algorithm is equally important. To complement our flow-level analysis, we analyze the traffic patterns of several mobile apps, finding that apps can be categorized as {"}short-flow dominated{"} or {"}long-flow dominated{"}. We then record and replay these patterns over emulatedWiFi and LTE links. We find that application performance has a similar dependence on the choice of networks as flow-level performance: An application dominated by short flows sees little gain from MPTCP, while an application with longer flows can benefit much more from MPTCP - if the application can pick the right network for the primary subflow and the right choice of MPTCP congestion control.",
keywords = "LTE, Mobile device, Multi-network, Multi-path TCP",
author = "Shuo Deng and Ravi Netravali and Anirudh Sivaraman and Hari Balakrishnan",
year = "2014",
month = "11",
day = "5",
doi = "10.1145/2663716.2663727",
language = "English (US)",
pages = "181--194",
booktitle = "IMC 2014 - Proceedings of the 2014 ACM",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - WiFi, LTE, or both? Measuring multi-homed wireless internet performance

AU - Deng, Shuo

AU - Netravali, Ravi

AU - Sivaraman, Anirudh

AU - Balakrishnan, Hari

PY - 2014/11/5

Y1 - 2014/11/5

N2 - Over the past two or three years, wireless cellular networks have become faster than before, most notably due to the deployment of LTE, HSPA+, and other similar networks. LTE throughputs can reach many megabits per second and can even rival WiFi throughputs in some locations. This paper addresses a fundamental question confronting transport and application-layer protocol designers: Which network should an application use? WiFi, LTE, or Multi-Path TCP (MPTCP) running over both? We compare LTE and WiFi for transfers of different sizes along both directions (i.e. the uplink and the downlink) using a crowdsourced mobile application run by 750 users over 180 days in 16 different countries. We find that LTE outperforms WiFi 40% of the time, which is a higher fraction than one might expect at first sight. We measure flow-level MPTCP performance and compare it with the performance of TCP running over exclusively WiFi or LTE in 20 different locations across 7 cities in the United States. For short flows, we find that MPTCP performs worse than regular TCP running over the faster link; further, selecting the correct network for the primary subflow in MPTCP is critical in achieving good performance. For long flows, however, selecting the proper MPTCP congestion control algorithm is equally important. To complement our flow-level analysis, we analyze the traffic patterns of several mobile apps, finding that apps can be categorized as "short-flow dominated" or "long-flow dominated". We then record and replay these patterns over emulatedWiFi and LTE links. We find that application performance has a similar dependence on the choice of networks as flow-level performance: An application dominated by short flows sees little gain from MPTCP, while an application with longer flows can benefit much more from MPTCP - if the application can pick the right network for the primary subflow and the right choice of MPTCP congestion control.

AB - Over the past two or three years, wireless cellular networks have become faster than before, most notably due to the deployment of LTE, HSPA+, and other similar networks. LTE throughputs can reach many megabits per second and can even rival WiFi throughputs in some locations. This paper addresses a fundamental question confronting transport and application-layer protocol designers: Which network should an application use? WiFi, LTE, or Multi-Path TCP (MPTCP) running over both? We compare LTE and WiFi for transfers of different sizes along both directions (i.e. the uplink and the downlink) using a crowdsourced mobile application run by 750 users over 180 days in 16 different countries. We find that LTE outperforms WiFi 40% of the time, which is a higher fraction than one might expect at first sight. We measure flow-level MPTCP performance and compare it with the performance of TCP running over exclusively WiFi or LTE in 20 different locations across 7 cities in the United States. For short flows, we find that MPTCP performs worse than regular TCP running over the faster link; further, selecting the correct network for the primary subflow in MPTCP is critical in achieving good performance. For long flows, however, selecting the proper MPTCP congestion control algorithm is equally important. To complement our flow-level analysis, we analyze the traffic patterns of several mobile apps, finding that apps can be categorized as "short-flow dominated" or "long-flow dominated". We then record and replay these patterns over emulatedWiFi and LTE links. We find that application performance has a similar dependence on the choice of networks as flow-level performance: An application dominated by short flows sees little gain from MPTCP, while an application with longer flows can benefit much more from MPTCP - if the application can pick the right network for the primary subflow and the right choice of MPTCP congestion control.

KW - LTE

KW - Mobile device

KW - Multi-network

KW - Multi-path TCP

UR - http://www.scopus.com/inward/record.url?scp=84910131969&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84910131969&partnerID=8YFLogxK

U2 - 10.1145/2663716.2663727

DO - 10.1145/2663716.2663727

M3 - Conference contribution

SP - 181

EP - 194

BT - IMC 2014 - Proceedings of the 2014 ACM

PB - Association for Computing Machinery

ER -