Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks

Shaojun Zou; Jiawei Huang; Yutao Zhou; Jianxin Wang; Tian He

doi:10.1109/ICDCS.2017.170

Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks

Shaojun Zou, Jiawei Huang, Yutao Zhou, Jianxin Wang, Tian He

Center for Transportation Studies

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

8 Scopus citations

Abstract

In data center networks, many network-intensive applications leverage large fan-in and many-to-one communication to achieve high performance. However, the special traffic patterns, such as micro-burst and high concurrency, easily cause TCP Incast problem and seriously degrade the application performance. To address the TCP Incast problem, we first reveal theoretically and empirically that alleviating packet burstiness is much more effective in reducing the Incast probability than controlling the congestion window. Inspired by the findings and insights from our experimental observations, we further propose a general supporting scheme Adaptive Pacing (AP), which dynamically adjusts burstiness according to the flow concurrency without any change on switch. Another feature of AP is its broad applicability. We integrate AP transparently into different TCP protocols (i.e., DCTCP, L2DCT and D2TCP). Through a series of large-scale NS2 simulations, we show that AP significantly reduces the Incast probability across different TCP protocols and the network goodput can be increased consistently by on average 7x under severe congestion.

Original language	English (US)
Title of host publication	Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017
Editors	Kisung Lee, Ling Liu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2119-2124
Number of pages	6
ISBN (Electronic)	9781538617915
DOIs	https://doi.org/10.1109/ICDCS.2017.170
State	Published - Jul 13 2017
Event	37th IEEE International Conference on Distributed Computing Systems, ICDCS 2017 - Atlanta, United States Duration: Jun 5 2017 → Jun 8 2017

Publication series

Name	Proceedings - International Conference on Distributed Computing Systems

Other

Other	37th IEEE International Conference on Distributed Computing Systems, ICDCS 2017
Country/Territory	United States
City	Atlanta
Period	6/5/17 → 6/8/17

Bibliographical note

Funding Information:
This work is supported by the National Natural Science Foundation of China (61572530, 61502539, 61402541, 61462007 and 61420106009) and the Next Generation Internet Innovation Foundation (Grant No. NGII201601130).

Publisher Copyright:
© 2017 IEEE.

Keywords

Congestion control
Data center
Incast
Pacing
TCP

Access

10.1109/ICDCS.2017.170

OpenUrl availability

Full text

Cite this

Zou, S., Huang, J., Zhou, Y., Wang, J., & He, T. (2017). Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks. In K. Lee, & L. Liu (Eds.), Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017 (pp. 2119-2124). Article 7980158 (Proceedings - International Conference on Distributed Computing Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDCS.2017.170

Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks. / Zou, Shaojun; Huang, Jiawei; Zhou, Yutao et al.
Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017. ed. / Kisung Lee; Ling Liu. Institute of Electrical and Electronics Engineers Inc., 2017. p. 2119-2124 7980158 (Proceedings - International Conference on Distributed Computing Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zou, S, Huang, J, Zhou, Y, Wang, J & He, T 2017, Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks. in K Lee & L Liu (eds), Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017., 7980158, Proceedings - International Conference on Distributed Computing Systems, Institute of Electrical and Electronics Engineers Inc., pp. 2119-2124, 37th IEEE International Conference on Distributed Computing Systems, ICDCS 2017, Atlanta, United States, 6/5/17. https://doi.org/10.1109/ICDCS.2017.170

Zou S, Huang J, Zhou Y, Wang J, He T. Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks. In Lee K, Liu L, editors, Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 2119-2124. 7980158. (Proceedings - International Conference on Distributed Computing Systems). doi: 10.1109/ICDCS.2017.170

Zou, Shaojun ; Huang, Jiawei ; Zhou, Yutao et al. / Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks. Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017. editor / Kisung Lee ; Ling Liu. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 2119-2124 (Proceedings - International Conference on Distributed Computing Systems).

@inproceedings{105f6221670341b2b0b2c918bd4238ef,

title = "Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks",

abstract = "In data center networks, many network-intensive applications leverage large fan-in and many-to-one communication to achieve high performance. However, the special traffic patterns, such as micro-burst and high concurrency, easily cause TCP Incast problem and seriously degrade the application performance. To address the TCP Incast problem, we first reveal theoretically and empirically that alleviating packet burstiness is much more effective in reducing the Incast probability than controlling the congestion window. Inspired by the findings and insights from our experimental observations, we further propose a general supporting scheme Adaptive Pacing (AP), which dynamically adjusts burstiness according to the flow concurrency without any change on switch. Another feature of AP is its broad applicability. We integrate AP transparently into different TCP protocols (i.e., DCTCP, L2DCT and D2TCP). Through a series of large-scale NS2 simulations, we show that AP significantly reduces the Incast probability across different TCP protocols and the network goodput can be increased consistently by on average 7x under severe congestion.",

keywords = "Congestion control, Data center, Incast, Pacing, TCP",

author = "Shaojun Zou and Jiawei Huang and Yutao Zhou and Jianxin Wang and Tian He",

note = "Funding Information: This work is supported by the National Natural Science Foundation of China (61572530, 61502539, 61402541, 61462007 and 61420106009) and the Next Generation Internet Innovation Foundation (Grant No. NGII201601130). Publisher Copyright: {\textcopyright} 2017 IEEE.; 37th IEEE International Conference on Distributed Computing Systems, ICDCS 2017 ; Conference date: 05-06-2017 Through 08-06-2017",

year = "2017",

month = jul,

day = "13",

doi = "10.1109/ICDCS.2017.170",

language = "English (US)",

series = "Proceedings - International Conference on Distributed Computing Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "2119--2124",

editor = "Kisung Lee and Ling Liu",

booktitle = "Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017",

}

TY - GEN

T1 - Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks

AU - Zou, Shaojun

AU - Huang, Jiawei

AU - Zhou, Yutao

AU - Wang, Jianxin

AU - He, Tian

N1 - Funding Information: This work is supported by the National Natural Science Foundation of China (61572530, 61502539, 61402541, 61462007 and 61420106009) and the Next Generation Internet Innovation Foundation (Grant No. NGII201601130). Publisher Copyright: © 2017 IEEE.

PY - 2017/7/13

Y1 - 2017/7/13

N2 - In data center networks, many network-intensive applications leverage large fan-in and many-to-one communication to achieve high performance. However, the special traffic patterns, such as micro-burst and high concurrency, easily cause TCP Incast problem and seriously degrade the application performance. To address the TCP Incast problem, we first reveal theoretically and empirically that alleviating packet burstiness is much more effective in reducing the Incast probability than controlling the congestion window. Inspired by the findings and insights from our experimental observations, we further propose a general supporting scheme Adaptive Pacing (AP), which dynamically adjusts burstiness according to the flow concurrency without any change on switch. Another feature of AP is its broad applicability. We integrate AP transparently into different TCP protocols (i.e., DCTCP, L2DCT and D2TCP). Through a series of large-scale NS2 simulations, we show that AP significantly reduces the Incast probability across different TCP protocols and the network goodput can be increased consistently by on average 7x under severe congestion.

AB - In data center networks, many network-intensive applications leverage large fan-in and many-to-one communication to achieve high performance. However, the special traffic patterns, such as micro-burst and high concurrency, easily cause TCP Incast problem and seriously degrade the application performance. To address the TCP Incast problem, we first reveal theoretically and empirically that alleviating packet burstiness is much more effective in reducing the Incast probability than controlling the congestion window. Inspired by the findings and insights from our experimental observations, we further propose a general supporting scheme Adaptive Pacing (AP), which dynamically adjusts burstiness according to the flow concurrency without any change on switch. Another feature of AP is its broad applicability. We integrate AP transparently into different TCP protocols (i.e., DCTCP, L2DCT and D2TCP). Through a series of large-scale NS2 simulations, we show that AP significantly reduces the Incast probability across different TCP protocols and the network goodput can be increased consistently by on average 7x under severe congestion.

KW - Congestion control

KW - Data center

KW - Incast

KW - Pacing

KW - TCP

UR - http://www.scopus.com/inward/record.url?scp=85027273246&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027273246&partnerID=8YFLogxK

U2 - 10.1109/ICDCS.2017.170

DO - 10.1109/ICDCS.2017.170

M3 - Conference contribution

AN - SCOPUS:85027273246

T3 - Proceedings - International Conference on Distributed Computing Systems

SP - 2119

EP - 2124

BT - Proceedings - IEEE 37th International Conference on Distributed Computing Systems, ICDCS 2017

A2 - Lee, Kisung

A2 - Liu, Ling

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 37th IEEE International Conference on Distributed Computing Systems, ICDCS 2017

Y2 - 5 June 2017 through 8 June 2017

ER -

Flow-Aware Adaptive Pacing to Mitigate TCP Incast in Data Center Networks

Abstract

Publication series

Other

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this