More Web Proxy on the site http://driver.im/

research-article

In-Network Neural Networks: Challenges and Opportunities for Innovation

Authors:

Marcelo C. Luizelli,

Ronaldo Canofre,

Arthur F. Lorenzon,

Fábio D. Rossi,

Weverton Cordeiro,

Oscar M. CaicedoAuthors Info & Claims

IEEE Network, Volume 35, Issue 6

Pages 68 - 74

https://doi.org/10.1109/MNET.101.2100098

Published: 01 November 2021 Publication History

Abstract

The quest for self-driving networks poses growing pressure to manage network events at a nano-second scale. In this article, we make a case for leveraging programmable forwarding planes to achieve self-driving networks and respond to their dynamism in real time by in-network intelligence and without performing traffic steering/mirroring to centralized management solutions (intelligent or not). We briefly cover throughout the article preliminary ideas in the in-network neural networks field and discuss the technical challenges of running machine learning techniques entirely in the forwarding plane. We also highlight potential use cases of having an autonomous intelligent network capable of self-adapting to dynamic network behavior changes with minimal to no human intervention, including smart network telemetry, smart traffic engineering, real-time flow classification, and network tomography. We close with a roadmap of research opportunities enabled by distributed in-network intelligence in programma-ble forwarding planes.

References

[1]

J. Zerwas et al., “Netboa: Self-Driving Network Benchmarking,” Proc. Wksp. Network Meets AI & ML, ACM, 2019, pp. 8–14.

[2]

G. Siracusano and R. Bifulco, “In-Network Neural Networks,” CoRR, vol. abs/1801.05731, 2018; http://arxiv.org/abs/1801.05731

[3]

A. Sapio et al., “Innetwork Computation Is a Dumb Idea Whose Time Has Come,” Proc. ACM HotNets '17, 2017, pp.150–56.

[4]

“Scaling Distributed Machine Learning with In-Network Aggregation,” Apr. 2021; https://www.usenix.org/confer-ence/nsdi21/presentation/sapio

[5]

Y. Li et al., “Accelerating Distributed Reinforcement Learning with In-Switch Computing,” Proc. lnt'l. Symp. Computer Architecture, ACM, 2019, p. 279–91.

[6]

D. Wu et al., “Accelerated Service Chaining on A Single Switch ASIC,” Proc. ACM HotNets'19, 2019, pp. 141–49.

[7]

Z. Xiong and N. Zilberman, “Do Switches Dream of Machine Learning? Toward In-Network Classification,” Proc. ACM HotNets'19, 2019, pp. 25–33.

[8]

D. Sanvito, G. Siracusano, and R. Bifulco, “Can the Network Be the Ai Accelerator?,” Proc. Wksp. In-Network Computing, 2018, pp. 20–25.

[9]

R. B. Basat et al., “Pint: Probabilistic In-Band Network Telem-etry,” ACM SIGCOMM 2020, 2020.

[10]

R. Hohemberger et al., “Orchestrating In-Band Data Plane Telemetry with Machine Learning,” IEEE Commun. Letters, vol. 23, no. 12, Dec 2019, pp. 2247–51.

[11]

F. Estrada-Solano, O. M. Caicedo, and N. L. S. Da Fonseca, “Nelly: Flow Detection Using Incremental Learning at the Server Side of SDN-Based Data Centers,” IEEE Trans. Indus-trial Informatics, vol. 16, no. 2, 2020, pp. 1362–72.

[12]

M. Hamdan et al., “Flow-Aware Elephant Flow Detection for Software-Defined Networks,” IEEE Access, vol. 8, 2020, pp. 72,585–97.

[13]

K.-F. Hsu et al., “Adaptive Weighted Traffic Splitting in Pro-grammable Data Planes,” ACM SOSR, 2020, pp. 103–09.

[14]

T. Guan et al., “Recursive Binary Neural Network Training Model for Efficient Usage of on-Chip Memory,” IEEE Trans. Circuits and Systems I, vol. 66, no. 7, 2019, pp. 2593–2605.

[15]

Q. Qin et al., “Line-Speed and Scalable Intrusion Detection at the Network Edge Via Federated Learning,” IFIP Networking, 2020, pp. 352–60.

Cited By

Bardhi EConti MLazzeretti R(2024)Is AI a Trick or T(h)reat for Securing Programmable Data Planes?IEEE Network: The Magazine of Global Internetworking10.1109/MNET.2024.345133038:6(146-152)Online publication date: 1-Nov-2024
https://dl.acm.org/doi/10.1109/MNET.2024.3451330
Parizotto RCoelho BNunes DHaque ISchaeffer-Filho A(2023)Offloading Machine Learning to Programmable Data Planes: A Systematic SurveyACM Computing Surveys10.1145/360515356:1(1-34)Online publication date: 26-Aug-2023
https://dl.acm.org/doi/10.1145/3605153
Liu WLiang CCui YCai JLuo J(2023)Programmable Data Plane Intelligence: Advances, Opportunities, and ChallengesIEEE Network: The Magazine of Global Internetworking10.1109/MNET.124.220011337:5(122-128)Online publication date: 1-Sep-2023
https://dl.acm.org/doi/10.1109/MNET.124.2200113

Index Terms

In-Network Neural Networks: Challenges and Opportunities for Innovation

Index terms have been assigned to the content through auto-classification.

Recommendations

Opportunities and Challenges of Software-Defined Mobile Networks in Network Security

Software-defined mobile network (SDMN) architecture integrates software-defined networks, network functions virtualization, and cloud computing principles in mobile networking environments to transform rigid and disparate legacy mobile networks into ...
Network for AI and AI for Network: Challenges and Opportunities for Learning-Oriented Networks
The “data pipe” model used by the existing Internet protocol stack is no longer ideal for many emerging applications, due to multimedia, multicast, mobility, machine learning, and network management challenges. A new learning-oriented ...
Opportunities and research challenges of hybrid software defined networks

Software Defined Networking (SDN) promises to ease design, operation and management of communication networks. However, SDN comes with its own set of challenges, including incremental deployability, robustness, and scalability. Those challenges make a ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image IEEE Network: The Magazine of Global Internetworking

IEEE Network: The Magazine of Global Internetworking Volume 35, Issue 6

November/December 2021

295 pages

ISSN:0890-8044

Issue’s Table of Contents

Publisher

IEEE Press

Publication History

Published: 01 November 2021

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

3
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 01 Mar 2025

Other Metrics

View Author Metrics

Citations

Cited By

Bardhi EConti MLazzeretti R(2024)Is AI a Trick or T(h)reat for Securing Programmable Data Planes?IEEE Network: The Magazine of Global Internetworking10.1109/MNET.2024.345133038:6(146-152)Online publication date: 1-Nov-2024
https://dl.acm.org/doi/10.1109/MNET.2024.3451330
Parizotto RCoelho BNunes DHaque ISchaeffer-Filho A(2023)Offloading Machine Learning to Programmable Data Planes: A Systematic SurveyACM Computing Surveys10.1145/360515356:1(1-34)Online publication date: 26-Aug-2023
https://dl.acm.org/doi/10.1145/3605153
Liu WLiang CCui YCai JLuo J(2023)Programmable Data Plane Intelligence: Advances, Opportunities, and ChallengesIEEE Network: The Magazine of Global Internetworking10.1109/MNET.124.220011337:5(122-128)Online publication date: 1-Sep-2023
https://dl.acm.org/doi/10.1109/MNET.124.2200113

View Options

View options

Figures

Tables

Media

View Issue’s Table of Contents