CLEO: Machine learning for ECMP

Heesang Jin; Minkoo Kang; Gyeongsik Yang; Chuck Yoo

doi:10.1145/3360468.3366768

CLEO: Machine learning for ECMP

Heesang Jin, Minkoo Kang, Gyeongsik Yang, Chuck Yoo

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

1 Citation (Scopus)

Abstract

In this paper, we propose CLEO, which is a machine learning approach to equal-cost multipath routing (ECMP) schemes to distribute and balance traffic. ECMP-based traffic load-balancing is widely practiced by datacenters, but hash collision resulting from skewed ECMP hashing makes it difficult to achieve the desired throughputs over paths. Various solutions have been proposed to overcome the performance degradation caused by hash collision, but most of these solutions require modifying packet headers or replacing switches. To solve this problem, CLEO builds a neural-network model that characterizes the ECMP scheme of a switch. The proof-of-concept evaluation shows that CLEO improves the root mean square error fourfold between the desired and real path throughputs.

Original language	English
Title of host publication	CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019
Publisher	Association for Computing Machinery, Inc
Pages	1-3
Number of pages	3
ISBN (Electronic)	9781450370066
DOIs	https://doi.org/10.1145/3360468.3366768
Publication status	Published - 2019 Dec 9
Event	15th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2019 - Part of CoNEXT 2019 - Orlando, United States Duration: 2019 Dec 9 → 2019 Dec 12

Publication series

Name	CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019

Conference

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

Bibliographical note

Funding Information:
We would like to thank the anonymous reviewers for their insightful comments. This work was supported by Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korean government (MSIT) (No. 2015-0-00288, Research of Network Virtualization Platform and Service for SDN 2.0 Realization, and No. 2015-0-00280, (SW Starlab) Next generation cloud infra-software toward the guarantee of performance and security SLA).

Publisher Copyright:
© 2019 held by the owner/author(s).

ASJC Scopus subject areas

Hardware and Architecture
Electrical and Electronic Engineering
Computer Networks and Communications

Access to Document

10.1145/3360468.3366768

Cite this

Jin, H., Kang, M., Yang, G., & Yoo, C. (2019). CLEO: Machine learning for ECMP. In CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019 (pp. 1-3). (CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019). Association for Computing Machinery, Inc. https://doi.org/10.1145/3360468.3366768

CLEO: Machine learning for ECMP. / Jin, Heesang; Kang, Minkoo; Yang, Gyeongsik et al.
CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019. Association for Computing Machinery, Inc, 2019. p. 1-3 (CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019).

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

Jin, H, Kang, M, Yang, G & Yoo, C 2019, CLEO: Machine learning for ECMP. in CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019. CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019, Association for Computing Machinery, Inc, pp. 1-3, 15th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2019 - Part of CoNEXT 2019, Orlando, United States, 19/12/9. https://doi.org/10.1145/3360468.3366768

Jin H, Kang M, Yang G, Yoo C. CLEO: Machine learning for ECMP. In CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019. Association for Computing Machinery, Inc. 2019. p. 1-3. (CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019). doi: 10.1145/3360468.3366768

Jin, Heesang ; Kang, Minkoo ; Yang, Gyeongsik et al. / CLEO : Machine learning for ECMP. CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019. Association for Computing Machinery, Inc, 2019. pp. 1-3 (CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019).

@inproceedings{6387068ec000426dade2eb20956129f7,

title = "CLEO: Machine learning for ECMP",

abstract = "In this paper, we propose CLEO, which is a machine learning approach to equal-cost multipath routing (ECMP) schemes to distribute and balance traffic. ECMP-based traffic load-balancing is widely practiced by datacenters, but hash collision resulting from skewed ECMP hashing makes it difficult to achieve the desired throughputs over paths. Various solutions have been proposed to overcome the performance degradation caused by hash collision, but most of these solutions require modifying packet headers or replacing switches. To solve this problem, CLEO builds a neural-network model that characterizes the ECMP scheme of a switch. The proof-of-concept evaluation shows that CLEO improves the root mean square error fourfold between the desired and real path throughputs.",

author = "Heesang Jin and Minkoo Kang and Gyeongsik Yang and Chuck Yoo",

note = "Funding Information: We would like to thank the anonymous reviewers for their insightful comments. This work was supported by Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korean government (MSIT) (No. 2015-0-00288, Research of Network Virtualization Platform and Service for SDN 2.0 Realization, and No. 2015-0-00280, (SW Starlab) Next generation cloud infra-software toward the guarantee of performance and security SLA). Publisher Copyright: {\textcopyright} 2019 held by the owner/author(s).; 15th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2019 - Part of CoNEXT 2019 ; Conference date: 09-12-2019 Through 12-12-2019",

year = "2019",

month = dec,

day = "9",

doi = "10.1145/3360468.3366768",

language = "English",

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

publisher = "Association for Computing Machinery, Inc",

pages = "1--3",

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

}

TY - GEN

T1 - CLEO

T2 - 15th International Conference on Emerging Networking EXperiments and Technologies, CoNEXT 2019 - Part of CoNEXT 2019

AU - Jin, Heesang

AU - Kang, Minkoo

AU - Yang, Gyeongsik

AU - Yoo, Chuck

N1 - Funding Information: We would like to thank the anonymous reviewers for their insightful comments. This work was supported by Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korean government (MSIT) (No. 2015-0-00288, Research of Network Virtualization Platform and Service for SDN 2.0 Realization, and No. 2015-0-00280, (SW Starlab) Next generation cloud infra-software toward the guarantee of performance and security SLA). Publisher Copyright: © 2019 held by the owner/author(s).

PY - 2019/12/9

Y1 - 2019/12/9

N2 - In this paper, we propose CLEO, which is a machine learning approach to equal-cost multipath routing (ECMP) schemes to distribute and balance traffic. ECMP-based traffic load-balancing is widely practiced by datacenters, but hash collision resulting from skewed ECMP hashing makes it difficult to achieve the desired throughputs over paths. Various solutions have been proposed to overcome the performance degradation caused by hash collision, but most of these solutions require modifying packet headers or replacing switches. To solve this problem, CLEO builds a neural-network model that characterizes the ECMP scheme of a switch. The proof-of-concept evaluation shows that CLEO improves the root mean square error fourfold between the desired and real path throughputs.

AB - In this paper, we propose CLEO, which is a machine learning approach to equal-cost multipath routing (ECMP) schemes to distribute and balance traffic. ECMP-based traffic load-balancing is widely practiced by datacenters, but hash collision resulting from skewed ECMP hashing makes it difficult to achieve the desired throughputs over paths. Various solutions have been proposed to overcome the performance degradation caused by hash collision, but most of these solutions require modifying packet headers or replacing switches. To solve this problem, CLEO builds a neural-network model that characterizes the ECMP scheme of a switch. The proof-of-concept evaluation shows that CLEO improves the root mean square error fourfold between the desired and real path throughputs.

UR - http://www.scopus.com/inward/record.url?scp=85077961343&partnerID=8YFLogxK

U2 - 10.1145/3360468.3366768

DO - 10.1145/3360468.3366768

M3 - Conference contribution

AN - SCOPUS:85077961343

T3 - CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019

SP - 1

EP - 3

BT - CoNEXT 2019 Companion - Proceedings of the 15th International Conference on Emerging Networking EXperiments and Technologies, Part of CoNEXT 2019

PB - Association for Computing Machinery, Inc

Y2 - 9 December 2019 through 12 December 2019

ER -

CLEO: Machine learning for ECMP

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this