Adaptive congestion control for unpredictable cellular networks

Yasir Zaki, Thomas Poestch, Jay Chen, Lakshminarayanan Subramanian, Carmelita Görg

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

Abstract

Legacy congestion controls including TCP and its variants are known to perform poorly over cellular networks due to highly variable capacities over short time scales, self inicted packet delays, and packet losses unrelated to congestion. To cope with these challenges, we present Verus, an end-to-end congestion control protocol that uses delay measurements to react quickly to the capacity changes in cellular networks without explicitly attempting to predict the cellular channel dynamics. The key idea of Verus is to continuously learn a delay profile that captures the relationship between end-to-end packet delay and outstanding window size over short epochs and uses this relationship to increment or decrement the window size based on the observed short-term packet delay variations. While the delay-based control is primarily for congestion avoidance, Verus uses standard TCP features including multiplicative decrease upon packet loss and slow start. Through a combination of simulations, empirical evaluations using cellular network traces, and real-world evaluations against standard TCP avors and state of the art protocols like Sprout, we show that Verus outperforms these protocols in cellular channels. In comparison to TCP Cubic, Verus achieves an order of magnitude (> 10x) reduction in delay over 3G and LTE networks while achieving comparable throughput (some- Times marginally higher). In comparison to Sprout, Verus achieves up to 30% higher throughput in rapidly changing cellular networks.

Original languageEnglish (US)
Title of host publicationSIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication
PublisherAssociation for Computing Machinery, Inc
Pages509-522
Number of pages14
ISBN (Print)9781450335423
DOIs
StatePublished - Aug 17 2015
EventACM Conference on Special Interest Group on Data Communication, SIGCOMM 2015 - London, United Kingdom
Duration: Aug 17 2015Aug 21 2015

Other

OtherACM Conference on Special Interest Group on Data Communication, SIGCOMM 2015
CountryUnited Kingdom
CityLondon
Period8/17/158/21/15

Fingerprint

Packet loss
Throughput
evaluation
simulation
time

Keywords

  • Cellular network
  • Congestion control
  • Delay-based
  • Transport protocol

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Signal Processing
  • Electrical and Electronic Engineering
  • Communication

Cite this

Zaki, Y., Poestch, T., Chen, J., Subramanian, L., & Görg, C. (2015). Adaptive congestion control for unpredictable cellular networks. In SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication (pp. 509-522). Association for Computing Machinery, Inc. https://doi.org/10.1145/2785956.2787498

Adaptive congestion control for unpredictable cellular networks. / Zaki, Yasir; Poestch, Thomas; Chen, Jay; Subramanian, Lakshminarayanan; Görg, Carmelita.

SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. Association for Computing Machinery, Inc, 2015. p. 509-522.

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

Zaki, Y, Poestch, T, Chen, J, Subramanian, L & Görg, C 2015, Adaptive congestion control for unpredictable cellular networks. in SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. Association for Computing Machinery, Inc, pp. 509-522, ACM Conference on Special Interest Group on Data Communication, SIGCOMM 2015, London, United Kingdom, 8/17/15. https://doi.org/10.1145/2785956.2787498
Zaki Y, Poestch T, Chen J, Subramanian L, Görg C. Adaptive congestion control for unpredictable cellular networks. In SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. Association for Computing Machinery, Inc. 2015. p. 509-522 https://doi.org/10.1145/2785956.2787498
Zaki, Yasir ; Poestch, Thomas ; Chen, Jay ; Subramanian, Lakshminarayanan ; Görg, Carmelita. / Adaptive congestion control for unpredictable cellular networks. SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication. Association for Computing Machinery, Inc, 2015. pp. 509-522
@inproceedings{9f4090bdfc8b4200b6ffe52fdf414563,
title = "Adaptive congestion control for unpredictable cellular networks",
abstract = "Legacy congestion controls including TCP and its variants are known to perform poorly over cellular networks due to highly variable capacities over short time scales, self inicted packet delays, and packet losses unrelated to congestion. To cope with these challenges, we present Verus, an end-to-end congestion control protocol that uses delay measurements to react quickly to the capacity changes in cellular networks without explicitly attempting to predict the cellular channel dynamics. The key idea of Verus is to continuously learn a delay profile that captures the relationship between end-to-end packet delay and outstanding window size over short epochs and uses this relationship to increment or decrement the window size based on the observed short-term packet delay variations. While the delay-based control is primarily for congestion avoidance, Verus uses standard TCP features including multiplicative decrease upon packet loss and slow start. Through a combination of simulations, empirical evaluations using cellular network traces, and real-world evaluations against standard TCP avors and state of the art protocols like Sprout, we show that Verus outperforms these protocols in cellular channels. In comparison to TCP Cubic, Verus achieves an order of magnitude (> 10x) reduction in delay over 3G and LTE networks while achieving comparable throughput (some- Times marginally higher). In comparison to Sprout, Verus achieves up to 30{\%} higher throughput in rapidly changing cellular networks.",
keywords = "Cellular network, Congestion control, Delay-based, Transport protocol",
author = "Yasir Zaki and Thomas Poestch and Jay Chen and Lakshminarayanan Subramanian and Carmelita G{\"o}rg",
year = "2015",
month = "8",
day = "17",
doi = "10.1145/2785956.2787498",
language = "English (US)",
isbn = "9781450335423",
pages = "509--522",
booktitle = "SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Adaptive congestion control for unpredictable cellular networks

AU - Zaki, Yasir

AU - Poestch, Thomas

AU - Chen, Jay

AU - Subramanian, Lakshminarayanan

AU - Görg, Carmelita

PY - 2015/8/17

Y1 - 2015/8/17

N2 - Legacy congestion controls including TCP and its variants are known to perform poorly over cellular networks due to highly variable capacities over short time scales, self inicted packet delays, and packet losses unrelated to congestion. To cope with these challenges, we present Verus, an end-to-end congestion control protocol that uses delay measurements to react quickly to the capacity changes in cellular networks without explicitly attempting to predict the cellular channel dynamics. The key idea of Verus is to continuously learn a delay profile that captures the relationship between end-to-end packet delay and outstanding window size over short epochs and uses this relationship to increment or decrement the window size based on the observed short-term packet delay variations. While the delay-based control is primarily for congestion avoidance, Verus uses standard TCP features including multiplicative decrease upon packet loss and slow start. Through a combination of simulations, empirical evaluations using cellular network traces, and real-world evaluations against standard TCP avors and state of the art protocols like Sprout, we show that Verus outperforms these protocols in cellular channels. In comparison to TCP Cubic, Verus achieves an order of magnitude (> 10x) reduction in delay over 3G and LTE networks while achieving comparable throughput (some- Times marginally higher). In comparison to Sprout, Verus achieves up to 30% higher throughput in rapidly changing cellular networks.

AB - Legacy congestion controls including TCP and its variants are known to perform poorly over cellular networks due to highly variable capacities over short time scales, self inicted packet delays, and packet losses unrelated to congestion. To cope with these challenges, we present Verus, an end-to-end congestion control protocol that uses delay measurements to react quickly to the capacity changes in cellular networks without explicitly attempting to predict the cellular channel dynamics. The key idea of Verus is to continuously learn a delay profile that captures the relationship between end-to-end packet delay and outstanding window size over short epochs and uses this relationship to increment or decrement the window size based on the observed short-term packet delay variations. While the delay-based control is primarily for congestion avoidance, Verus uses standard TCP features including multiplicative decrease upon packet loss and slow start. Through a combination of simulations, empirical evaluations using cellular network traces, and real-world evaluations against standard TCP avors and state of the art protocols like Sprout, we show that Verus outperforms these protocols in cellular channels. In comparison to TCP Cubic, Verus achieves an order of magnitude (> 10x) reduction in delay over 3G and LTE networks while achieving comparable throughput (some- Times marginally higher). In comparison to Sprout, Verus achieves up to 30% higher throughput in rapidly changing cellular networks.

KW - Cellular network

KW - Congestion control

KW - Delay-based

KW - Transport protocol

UR - http://www.scopus.com/inward/record.url?scp=84962259677&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962259677&partnerID=8YFLogxK

U2 - 10.1145/2785956.2787498

DO - 10.1145/2785956.2787498

M3 - Conference contribution

SN - 9781450335423

SP - 509

EP - 522

BT - SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication

PB - Association for Computing Machinery, Inc

ER -