Fine-grained queue measurement in the data plane

Xiaoqi Chen; Shir  Landau Feibish; Yaron Koral; Jennifer Rexford; Ori Rottenstreich; Steven A. Monetti; Tzuu Yi Wang

doi:10.1145/3359989.3365408

Fine-grained queue measurement in the data plane

Xiaoqi Chen, Shir Landau Feibish, Yaron Koral, Jennifer Rexford, Ori Rottenstreich, Steven A. Monetti, Tzuu Yi Wang

Computer Science discipline

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

Abstract

Short-lived surges in traffic can cause periods of high queue utilization, leading to packet loss and delay. To diagnose and alleviate performance problems, networks need support for real-time, fine-grained queue measurement. By identifying the flows that contribute significantly to queue build-up directly in the data plane, switches can make targeted decisions to mark, drop, or reroute these flows in real time. However, collecting fine-grained queue statistics is challenging even with modern programmable switch hardware, due to limited memory and processing resources in the data plane. We present ConQuest, a compact data structure that identifies the flows making a significant contribution to the queue. ConQuest operates entirely in the data plane, while working within the hardware constraints of programmable switches. Additionally, we show how to measure queues in legacy devices through link tapping and an off-path switch running ConQuest. Simulations show that ConQuest can identify contributing flows with 90% precision on a 1 ms timescale, using less than 65 KB of memory. Experiments with our Barefoot Tofino prototype show that ConQuest-enabled active queue management reduces flow-completion time.

Original language	English
Title of host publication	CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies
Publisher	Association for Computing Machinery, Inc
Pages	15-29
Number of pages	15
ISBN (Electronic)	9781450369985
DOIs	https://doi.org/10.1145/3359989.3365408
State	Published - 3 Dec 2019
Event	15th ACM International Conference on Emerging Networking Experiments and Technologies, CoNEXT 2019 - Orlando, United States Duration: 9 Dec 2019 → 12 Dec 2019

Publication series

Name	CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies

Conference

Conference	15th ACM International Conference on Emerging Networking Experiments and Technologies, CoNEXT 2019
Country/Territory	United States
City	Orlando
Period	9/12/19 → 12/12/19

Bibliographical note

Funding Information:
Number of concurrent register memory accesses per stage supported by switch

Publisher Copyright:
© 2019 ACM.

Keywords

Network monitoring
P4
Queue measurement
SDN

Access to Document

10.1145/3359989.3365408

Cite this

Chen, X., Landau Feibish, S., Koral, Y., Rexford, J., Rottenstreich, O., Monetti, S. A., & Wang, T. Y. (2019). Fine-grained queue measurement in the data plane. In CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies (pp. 15-29). (CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies). Association for Computing Machinery, Inc. https://doi.org/10.1145/3359989.3365408

Chen, Xiaoqi ; Landau Feibish, Shir ; Koral, Yaron et al. / Fine-grained queue measurement in the data plane. CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies. Association for Computing Machinery, Inc, 2019. pp. 15-29 (CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies).

@inproceedings{fff3511ebcdc4ba2b4cbc96c0989ba98,

title = "Fine-grained queue measurement in the data plane",

abstract = "Short-lived surges in traffic can cause periods of high queue utilization, leading to packet loss and delay. To diagnose and alleviate performance problems, networks need support for real-time, fine-grained queue measurement. By identifying the flows that contribute significantly to queue build-up directly in the data plane, switches can make targeted decisions to mark, drop, or reroute these flows in real time. However, collecting fine-grained queue statistics is challenging even with modern programmable switch hardware, due to limited memory and processing resources in the data plane. We present ConQuest, a compact data structure that identifies the flows making a significant contribution to the queue. ConQuest operates entirely in the data plane, while working within the hardware constraints of programmable switches. Additionally, we show how to measure queues in legacy devices through link tapping and an off-path switch running ConQuest. Simulations show that ConQuest can identify contributing flows with 90% precision on a 1 ms timescale, using less than 65 KB of memory. Experiments with our Barefoot Tofino prototype show that ConQuest-enabled active queue management reduces flow-completion time.",

keywords = "Network monitoring, P4, Queue measurement, SDN",

author = "Xiaoqi Chen and {Landau Feibish}, Shir and Yaron Koral and Jennifer Rexford and Ori Rottenstreich and Monetti, {Steven A.} and Wang, {Tzuu Yi}",

note = "Publisher Copyright: {\textcopyright} 2019 ACM.; 15th ACM International Conference on Emerging Networking Experiments and Technologies, CoNEXT 2019 ; Conference date: 09-12-2019 Through 12-12-2019",

year = "2019",

month = dec,

day = "3",

doi = "10.1145/3359989.3365408",

language = "אנגלית",

series = "CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies",

publisher = "Association for Computing Machinery, Inc",

pages = "15--29",

booktitle = "CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies",

}

Chen, X, Landau Feibish, S, Koral, Y, Rexford, J, Rottenstreich, O, Monetti, SA & Wang, TY 2019, Fine-grained queue measurement in the data plane. in CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies. CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies, Association for Computing Machinery, Inc, pp. 15-29, 15th ACM International Conference on Emerging Networking Experiments and Technologies, CoNEXT 2019, Orlando, United States, 9/12/19. https://doi.org/10.1145/3359989.3365408

Fine-grained queue measurement in the data plane. / Chen, Xiaoqi; Landau Feibish, Shir ; Koral, Yaron et al.
CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies. Association for Computing Machinery, Inc, 2019. p. 15-29 (CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies).

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

TY - GEN

T1 - Fine-grained queue measurement in the data plane

AU - Chen, Xiaoqi

AU - Landau Feibish, Shir

AU - Koral, Yaron

AU - Rexford, Jennifer

AU - Rottenstreich, Ori

AU - Monetti, Steven A.

AU - Wang, Tzuu Yi

PY - 2019/12/3

Y1 - 2019/12/3

N2 - Short-lived surges in traffic can cause periods of high queue utilization, leading to packet loss and delay. To diagnose and alleviate performance problems, networks need support for real-time, fine-grained queue measurement. By identifying the flows that contribute significantly to queue build-up directly in the data plane, switches can make targeted decisions to mark, drop, or reroute these flows in real time. However, collecting fine-grained queue statistics is challenging even with modern programmable switch hardware, due to limited memory and processing resources in the data plane. We present ConQuest, a compact data structure that identifies the flows making a significant contribution to the queue. ConQuest operates entirely in the data plane, while working within the hardware constraints of programmable switches. Additionally, we show how to measure queues in legacy devices through link tapping and an off-path switch running ConQuest. Simulations show that ConQuest can identify contributing flows with 90% precision on a 1 ms timescale, using less than 65 KB of memory. Experiments with our Barefoot Tofino prototype show that ConQuest-enabled active queue management reduces flow-completion time.

AB - Short-lived surges in traffic can cause periods of high queue utilization, leading to packet loss and delay. To diagnose and alleviate performance problems, networks need support for real-time, fine-grained queue measurement. By identifying the flows that contribute significantly to queue build-up directly in the data plane, switches can make targeted decisions to mark, drop, or reroute these flows in real time. However, collecting fine-grained queue statistics is challenging even with modern programmable switch hardware, due to limited memory and processing resources in the data plane. We present ConQuest, a compact data structure that identifies the flows making a significant contribution to the queue. ConQuest operates entirely in the data plane, while working within the hardware constraints of programmable switches. Additionally, we show how to measure queues in legacy devices through link tapping and an off-path switch running ConQuest. Simulations show that ConQuest can identify contributing flows with 90% precision on a 1 ms timescale, using less than 65 KB of memory. Experiments with our Barefoot Tofino prototype show that ConQuest-enabled active queue management reduces flow-completion time.

KW - Network monitoring

KW - P4

KW - Queue measurement

KW - SDN

UR - http://www.scopus.com/inward/record.url?scp=85077229912&partnerID=8YFLogxK

U2 - 10.1145/3359989.3365408

DO - 10.1145/3359989.3365408

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85077229912

T3 - CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies

SP - 15

EP - 29

BT - CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies

PB - Association for Computing Machinery, Inc

T2 - 15th ACM International Conference on Emerging Networking Experiments and Technologies, CoNEXT 2019

Y2 - 9 December 2019 through 12 December 2019

ER -

Chen X, Landau Feibish S, Koral Y, Rexford J, Rottenstreich O, Monetti SA et al. Fine-grained queue measurement in the data plane. In CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies. Association for Computing Machinery, Inc. 2019. p. 15-29. (CoNEXT 2019 - Proceedings of the 15th International Conference on Emerging Networking Experiments and Technologies). doi: 10.1145/3359989.3365408

Fine-grained queue measurement in the data plane

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this