TM3: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking

Feng Qian; Vijay Gopalakrishnan; Emir Halepovic; Subhabrata Sen; Oliver Spatscheck

doi:10.1145/2716281.2836088

TM³: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking

Feng Qian, Vijay Gopalakrishnan, Emir Halepovic, Subhabrata Sen, Oliver Spatscheck

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

23 Scopus citations

Abstract

A primary design decision in HTTP/2, the successor of HTTP/1.1, is object multiplexing. While multiplexing improves web performance in many scenarios, it still has several drawbacks due to complex cross-layer interactions. In this paper, we propose a novel multiplexing architecture called TM³ that overcomes many of these limitations. TM³ strategically leverages multiple concurrent multiplexing pipes in a transparent manner, and eliminates various types of head-of-line blocking that can severely impact user experience. TM³ works beyond HTTP over TCP and applies to a wide range of application and transport protocols. Extensive evaluations on LTE and wired networks show that TM³ substantially improves performance e.g., reduces web page load time by an average of 24% compared to SPDY, which is the basis for HTTP/2. For lossy links and concurrent transfers, the improvements are more pronounced: compared to SPDY, TM³ achieves up to 42% of average PLT reduction under losses and up to 90% if concurrent transfers exist.

Original language	English (US)
Title of host publication	Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015
Publisher	Association for Computing Machinery, Inc
ISBN (Electronic)	9781450334129
DOIs	https://doi.org/10.1145/2716281.2836088
State	Published - Dec 1 2015
Externally published	Yes
Event	11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015 - Heidelberg, Germany Duration: Dec 1 2015 → Dec 4 2015

Publication series

Name	Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015

Other

Other	11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015
Country/Territory	Germany
City	Heidelberg
Period	12/1/15 → 12/4/15

Bibliographical note

Publisher Copyright:
© 2015 ACM.

Access

10.1145/2716281.2836088

OpenUrl availability

Full text

Cite this

Qian, F., Gopalakrishnan, V., Halepovic, E., Sen, S., & Spatscheck, O. (2015). TM³: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking. In Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015 Article 2836088 (Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015). Association for Computing Machinery, Inc. https://doi.org/10.1145/2716281.2836088

TM³: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking. / Qian, Feng; Gopalakrishnan, Vijay; Halepovic, Emir et al.
Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015. Association for Computing Machinery, Inc, 2015. 2836088 (Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015).

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

Qian, F, Gopalakrishnan, V, Halepovic, E, Sen, S & Spatscheck, O 2015, TM³: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking. in Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015., 2836088, Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015, Association for Computing Machinery, Inc, 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015, Heidelberg, Germany, 12/1/15. https://doi.org/10.1145/2716281.2836088

Qian F, Gopalakrishnan V, Halepovic E, Sen S, Spatscheck O. TM³: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking. In Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015. Association for Computing Machinery, Inc. 2015. 2836088. (Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015). doi: 10.1145/2716281.2836088

Qian, Feng ; Gopalakrishnan, Vijay ; Halepovic, Emir et al. / TM³ : Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking. Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015. Association for Computing Machinery, Inc, 2015. (Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015).

@inproceedings{e336121681b14c4bae117458a45ccbd9,

title = "TM3: Flexible transport-layer multi-pipe multiplexing middlebox without head-of-line blocking",

abstract = "A primary design decision in HTTP/2, the successor of HTTP/1.1, is object multiplexing. While multiplexing improves web performance in many scenarios, it still has several drawbacks due to complex cross-layer interactions. In this paper, we propose a novel multiplexing architecture called TM3 that overcomes many of these limitations. TM3 strategically leverages multiple concurrent multiplexing pipes in a transparent manner, and eliminates various types of head-of-line blocking that can severely impact user experience. TM3 works beyond HTTP over TCP and applies to a wide range of application and transport protocols. Extensive evaluations on LTE and wired networks show that TM3 substantially improves performance e.g., reduces web page load time by an average of 24% compared to SPDY, which is the basis for HTTP/2. For lossy links and concurrent transfers, the improvements are more pronounced: compared to SPDY, TM3 achieves up to 42% of average PLT reduction under losses and up to 90% if concurrent transfers exist.",

author = "Feng Qian and Vijay Gopalakrishnan and Emir Halepovic and Subhabrata Sen and Oliver Spatscheck",

note = "Publisher Copyright: {\textcopyright} 2015 ACM.; 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015 ; Conference date: 01-12-2015 Through 04-12-2015",

year = "2015",

month = dec,

day = "1",

doi = "10.1145/2716281.2836088",

language = "English (US)",

series = "Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015",

publisher = "Association for Computing Machinery, Inc",

booktitle = "Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015",

}

TY - GEN

T1 - TM3

T2 - 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015

AU - Qian, Feng

AU - Gopalakrishnan, Vijay

AU - Halepovic, Emir

AU - Sen, Subhabrata

AU - Spatscheck, Oliver

PY - 2015/12/1

Y1 - 2015/12/1

N2 - A primary design decision in HTTP/2, the successor of HTTP/1.1, is object multiplexing. While multiplexing improves web performance in many scenarios, it still has several drawbacks due to complex cross-layer interactions. In this paper, we propose a novel multiplexing architecture called TM3 that overcomes many of these limitations. TM3 strategically leverages multiple concurrent multiplexing pipes in a transparent manner, and eliminates various types of head-of-line blocking that can severely impact user experience. TM3 works beyond HTTP over TCP and applies to a wide range of application and transport protocols. Extensive evaluations on LTE and wired networks show that TM3 substantially improves performance e.g., reduces web page load time by an average of 24% compared to SPDY, which is the basis for HTTP/2. For lossy links and concurrent transfers, the improvements are more pronounced: compared to SPDY, TM3 achieves up to 42% of average PLT reduction under losses and up to 90% if concurrent transfers exist.

AB - A primary design decision in HTTP/2, the successor of HTTP/1.1, is object multiplexing. While multiplexing improves web performance in many scenarios, it still has several drawbacks due to complex cross-layer interactions. In this paper, we propose a novel multiplexing architecture called TM3 that overcomes many of these limitations. TM3 strategically leverages multiple concurrent multiplexing pipes in a transparent manner, and eliminates various types of head-of-line blocking that can severely impact user experience. TM3 works beyond HTTP over TCP and applies to a wide range of application and transport protocols. Extensive evaluations on LTE and wired networks show that TM3 substantially improves performance e.g., reduces web page load time by an average of 24% compared to SPDY, which is the basis for HTTP/2. For lossy links and concurrent transfers, the improvements are more pronounced: compared to SPDY, TM3 achieves up to 42% of average PLT reduction under losses and up to 90% if concurrent transfers exist.

UR - http://www.scopus.com/inward/record.url?scp=84994161453&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84994161453&partnerID=8YFLogxK

U2 - 10.1145/2716281.2836088

DO - 10.1145/2716281.2836088

M3 - Conference contribution

AN - SCOPUS:84994161453

T3 - Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015

BT - Proceedings of the 11th ACM Conference on Emerging Networking Experiments and Technologies, CoNEXT 2015

PB - Association for Computing Machinery, Inc

Y2 - 1 December 2015 through 4 December 2015

ER -