A Machine Learning-Based Toolbox for P4 Programmable Data-Planes
Pages 4450 - 4465
Abstract
Intelligent data-planes (IDPs) can enhance network service performance and adaptation speed by executing one or more machine learning (ML) models directly on the served flows. The real-time ML inference enables line-speed decision-making for some traffic management functionalities. Due to the inherent scarcity of both the computational and memory resources and the strict high-speed per-packet processing demands, existing IDP deployments either realize only a limited set of ML models such as decision trees, or require substantial modifications in the switch hardware. In this paper, we propose INQ-MLT, a novel ML-based management toolbox to address the aforementioned limitations. INQ-MLT delegates the task of training various ML models to the control-plane. The latter adopts a tailored quantization-aware training process to compensate for the effect of precision loss resulting from quantization. The toolbox then employs a quantization mechanism to transform the trained ML model parameters (e.g., weights and activations) from floating-point representations to compact low-precision fixed integer values that can be easily processed and stored in the data-plane. Finally, the trained model is deployed into the IDP pipeline by restricting all its inference operations to basic arithmetic operations. To analyze the performance of INQ-MLT, we quantify the accuracy loss resulting from the quantization step through rigorous theoretical analysis. A proof-of-concept implementation of the proposed toolbox is developed using P4-based software switches. Experiments on two use-cases demonstrate that the deployed quantized models have almost no loss of accuracy when compared to their floating-point counterparts.
References
[1]
M. Abbasi, A. Shahraki, and A. Taherkordi, “Deep learning for network traffic monitoring and analysis (NTMA): A survey,” Comput. Commun., vol. 170, pp. 19–41, Mar. 2021.
[2]
P. Poullie, T. Bocek, and B. Stiller, “A survey of the state-of-the-art in fair multi-resource allocations for data centers,” IEEE Trans. Netw. Service Manag., vol. 15, no. 1, pp. 169–183, Mar. 2017.
[3]
Y. Yu et al., “Fault management in software-defined networking: A survey,” IEEE Commun. Surveys Tuts., vol. 21, no. 1, pp. 349–392, 1st Quart., 2019.
[4]
R. Boutaba et al., “A comprehensive survey on machine learning for networking: Evolution, applications and research opportunities,” J. Internet Services Appl., vol. 9, no. 1, pp. 1–99, 2018.
[5]
D. Kreutz, F. M. V. Ramos, P. E. Verissimo, C. E. Rothenberg, S. Azodolmolky, and S. Uhlig, “Software-defined networking: A comprehensive survey,” Proc. IEEE, vol. 103, no. 1, pp. 14–76, Jan. 2015.
[6]
Y.-C. Wang and S.-Y. You, “An efficient route management framework for load balance and overhead reduction in SDN-based data center networks,” IEEE Trans. Netw. Service Manag., vol. 15, no. 4, pp. 1422–1434, Dec. 2018.
[7]
P. Bosshart et al. “P4: Programming protocol-independent packet processors,” ACM SIGCOMM Comput. Commun. Rev., vol. 44, no. 3, pp. 87–95, 2014.
[8]
A. Liatifis, P. Sarigiannidis, V. Argyriou, and T. Lagkas, “Advancing SDN from openflow to P4: A survey,” ACM Comput. Surveys, vol. 55, no. 9, pp. 1–37, 2023.
[9]
N. McKeown, “Pisa: Protocol independent switch architecture,” in P4 Workshop, pp. 1–22, 2015.
[10]
“v1model.” Accessed: Jan. 14, 2024. [Online]. Available: https://github.com/p4lang/p4c/blob/main/p4include/v1model.p4
[11]
(P4.org Architect. Working Group, Santa Clara, CA, USA). P416 Portable Switch Architecture (PSA). Accessed: Jan. 14, 2024. [Online]. Available: https://p4.org/p4-spec/docs/PSA-v1.2.html
[12]
(P.a Work. Group, Stanford, CA, USA). Initial Draft Specification for P4 Runtime, P4 Language Consortium. Oct. 2017. [Online]. Available: https://p4.org/p4-runtime/
[13]
T. Osi´nski, H. Tarasiuk, P. Chaignon, and M. Kossakowski, “A runtimeenabled p4 extension to the open vSwitch packet processing pipeline,” IEEE Trans. Netw. Service Manag., vol. 18, no. 3, pp. 2832–2845, Sep. 2021.
[14]
R. Parizotto, B. L. Coelho, D. C. Nunes, I. Haque, and A. Schaeffer-Filho, “Offloading machine learning to programmable data planes: A systematic survey,” ACM Comput. Surveys, vol. 36, no. 1, pp. 1–34, 2023.
[15]
B. M. Xavier, R. S. Guimarães, G. Comarela, and M. Martinello, “MAP4: A pragmatic framework for in-network machine learning traffic classification,” IEEE Trans. Netw. Service Manag., vol. 19, no. 4, pp. 4176–4188, Dec. 2022.
[16]
B. Coelho and A. Schaeffer-Filho, “Backorders: Using random forests to detect DDoS attacks in programmable data planes,” in Proc. 5th Int. Workshop P4 Eur., 2022, pp. 1–7.
[17]
K. Zhang, N. Samaan, and A. Karmouch, “A two-stage confidence-based intrusion detection system in programmable data-planes,” in Proc. IEEE Global Commun. Conf. GLOBECOM, 2023, pp. 1–6.
[18]
X. Zhang, L. Cui, F. P. Tso, and W. Jia, “pHeavy: Predicting heavy flows in the programmable data plane,” IEEE Trans. Netw. Service Manag., vol. 18, no. 4, pp. 4353–4364, Dec. 2021.
[19]
C. Zheng, B. Rienecker, and N. Zilberman, “QCMP: Load Balancing via In-network Reinforcement Learning,” in Proc. 2nd ACM SIGCOMM Workshop Future Internet Rout. Address., 2023, pp. 35–40.
[20]
C. Busse-Grawitz, R. Meier, A. Dietmüller, T. Bühler, and L. Vanbever, “pForest: In-network inference with random forests,” 2019, arXiv:1909.05680.
[21]
G. Xie et al., “Empowering in-network classification in programmable switches by binary decision tree and knowledge distillation,” IEEE/ACM Trans. Netw., vol. 32, no. 1, pp. 382–395, Feb. 2024.
[22]
Z. Xiong and N. Zilberman, “Do switches dream of machine learning? Toward in-network classification,” in Proc. 18th ACM Workshop Hot Topics Netw., 2019, pp. 25–33.
[23]
G. Siracusano et al., “Re-architecting traffic analysis with neural network interface cards,” in Proc. 19th USENIX Symp. Netw. Syst. Design Implement. (NSDI 22), 2022, pp. 513–533.
[24]
Q. Qin, K. Poularakis, K. K. Leung, and L. Tassiulas, “Line-speed and scalable intrusion detection at the network edge via federated learning,” in Proc. IFIP Netw. Conf. (Networking), 2020, pp. 352–360.
[25]
D. C. Li, M. R. Maulana, and L.-D. Chou, “NNSplit-SøREN: Supporting the model implementation of large neural networks in a programmable data plane,” Comput. Netw., vol. 222, Feb. 2023, Art. no.
[26]
Z. Zhong et al., “IOI: In-network optical inference,” in Proc. ACM SIGCOMM Workshop Opt. Syst., 2021, pp. 18–22.
[27]
T. Swamy, A. Rucker, M. Shahbaz, I. Gaur, and K. Olukotun, “Taurus: A data plane architecture for per-packet ML,” in Proc. 27th ACM Int. Conf. Architect. Support Program. Lang. Oper. Syst., 2022, pp. 1099–1114.
[28]
M. Jose, K. Lazri, J. François, and O. Festor, “Stateful InREC: Stateful in-network real number computation with recursive functions,” IEEE Trans. Netw. Service Manag., vol. 20, no. 1, pp. 830–845, Mar. 2023.
[29]
P. Cui et al., “Enabling in-network floating-point arithmetic for efficient computation offloading,” IEEE Trans. Parallel Distrib. Syst., vol. 33, no. 12, pp. 4918–4934, Dec. 2022.
[30]
K. Zhang, N. Samaan, and A. Karmouch, “An intelligent data-plane with a quantized ML model for traffic management,” in Proc. IEEE/IFIP Netw. Oper. Manag. Symp. (NOMS), 2023, pp. 1–9.
[31]
“Behavioral model version 2.” Accessed: May 22, 2022. [Online]. Available: https://github.com/p4lang/behavioral-model
[32]
J.-H. Lee and K. Singh, “SwitchTree: In-network computing and traffic analyses with random forests,” Neural Comput. Appl., 2020, to be Published.
[33]
C. Zheng et al., “IIsy: Hybrid in-network classification using programmable switches,” IEEE/ACM Trans. Netw., early access, Feb. 16, 2024. 10.1109/TNET.2024.3364757.
[34]
C. Zheng and N. Zilberman, “Planter: Seeding trees within switches,” in Proc. Poster Demo Sessions (SIGCOMM ‘21), New York, NY, USA, 2021, pp. 12–14.
[35]
C. Zheng et al., “Planter: Rapid prototyping of in-network machine learning inference,” ACM SIGCOMM Comput. Commun. Rev., 2024, to be Published.
[36]
G. Zhou, Z. Liu, C. Fu, Q. Li, and K. Xu, “An efficient design of intelligent network data plane,” in Proc. 32nd USENIX Security Symp. (USENIX Security 23), 2023, pp. 1–18.
[37]
C. Zheng, X. Hong, D. Ding, S. Vargaftik, Y. Ben-Itzhak, and N. Zilberman, “In-network machine learning using programmable network devices: A survey,” IEEE Commun. Surveys Tuts., early access, Dec. 19, 2023. 10.1109/COMST.2023.3344351.
[38]
T. Swamy, A. Zulfiqar, L. Nardi, M. Shahbaz, and K. Olukotun, “Homunculus: Auto-generating efficient data-plane ML pipelines for datacenter networks,” in Proc. 28th ACM Int. Conf. Architect. Support Program. Lang. Oper. Syst., 2023, pp. 329–342.
[39]
A. T.-J. Akem, M. Gucciardo, and M. Fiore, “Flowrest: Practical flowlevel inference in programmable switches with random forests,” in Proc. INFOCOM IEEE Conf. Comput. Commun., 2023, pp. 1–10.
[40]
E. Papadogiannaki and S. Ioannidis, “A survey on encrypted network traffic analysis applications, techniques, and countermeasures,” ACM Comput. Surveys, vol. 54, no. 6, pp. 1–35, 2021.
[41]
S. Rezaei and X. Liu, “Deep learning for encrypted traffic classification: An overview,” IEEE Commun. Mag., vol. 57, no. 5, pp. 76–81, May 2019.
[42]
T. T. T. Nguyen and G. Armitage, “A survey of techniques for Internet traffic classification using machine learning,” IEEE Commun. Surveys Tuts., vol. 10, no. 4, pp. 56–76, 4th Quart., 2008.
[43]
C. Xu, S. Chen, J. Su, S.-M. Yiu, and L. C. K. Hui, “A survey on regular expression matching for deep packet inspection: Applications, algorithms, and hardware platforms,” IEEE Commun. Surveys Tuts., vol. 18, no. 4, pp. 2991–3029, 4th Quart., 2016.
[44]
J. Hypolite, J. Sonchack, S. Hershkop, N. Dautenhahn, A. DeHon, and J. M. Smith, “Deepmatch: Practical deep packet inspection in the data plane using network processors,” in Proc. 16th Int. Conf. Emerg. Netw. Exp. Technol., 2020, pp. 336–350.
[45]
C. Liu, L. He, G. Xiong, Z. Cao, and Z. Li, “FS-Net: A flow sequence network for encrypted traffic classification,” in Proc. IEEE INFOCOM Conf. Comput. Commun., 2019, pp. 1171–1179.
[46]
T. Shapira and Y. Shavitt, “FlowPic: A generic representation for encrypted traffic classification and applications identification,” IEEE Trans. Netw. Service Manag., vol. 18, no. 2, pp. 1218–1232, Jun. 2021.
[47]
T.-L. Huoh, Y. Luo, P. Li, and T. Zhang, “Flow-based encrypted network traffic classification with graph neural networks,” IEEE Trans. Netw. Service Manag., vol. 20, no. 2, pp. 1224–1237, Jun. 2023.
[48]
Y. Wang, H. He, Y. Lai, and A. X. Liu, “A two-phase approach to fast and accurate classification of encrypted traffic,” IEEE/ACM Trans. Netw., vol. 31, no. 3, pp. 1071–1086, Jun. 2023.
[49]
X. Yun, Y. Wang, Y. Zhang, C. Zhao, and Z. Zhao, “Encrypted TLS traffic classification on cloud platforms,” IEEE/ACM Trans. Netw., vol. 31, no. 1, pp. 164–177, Feb. 2023.
[50]
R. Doriguzzi-Corin, S. Millar, S. Scott-Hayward, J. M.-d. Rincon, and D. Siracusa, “Lucid: A practical, lightweight deep learning solution for DDoS attack detection,” IEEE Trans. Netw. Service Manag., vol. 17, no. 2, pp. 876–889, Jun. 2020.
[51]
R. Vinayakumar, M. Alazab, K. P. Soman, P. Poornachandran, A. Al-Nemrat, and S. Venkatraman, “Deep learning approach for intelligent intrusion detection system,” IEEE Access, vol. 7, pp. 41525–41550, 2019.
[52]
L. Yang and A. Shami, “A transfer learning and optimized CNN based intrusion detection system for Internet of Vehicles,” in Proc. IEEE Int. Conf. Commun. (ICC), 2022, pp. 2774–2779.
[53]
S. Longari, D. H. Nova Valcarcel, M. Zago, M. Carminati, and S. Zanero, “CANnolo: An anomaly detection system based on LSTM autoencoders for controller area network,” IEEE Trans. Netw. Service Manag., vol. 18, no. 2, pp. 1913–1924, Jun. 2021.
[54]
I. Akbari et al., “A look behind the curtain: Traffic classification in an increasingly encrypted Web,” Proc. ACM Meas. Anal. Comput. Syst., vol. 5, no. 1, pp. 1–26, 2021.
[55]
S. Gupta, D. Gosain, M. Kwon, and H. B. Acharya, “Deep4r: Deep packet inspection in P4 using packet recirculation,” in Proc. IEEE INFOCOM Conf. Comput. Commun., 2023, pp. 1–10.
[56]
L. Tan et al., “In-band network telemetry: A survey,” Comput. Netw., vol. 186, Feb. 2021, Art. no.
[57]
R. B. Basat, S. Ramanathan, Y. Li, G. Antichi, M. Yu, and M. Mitzenmacher, “Pint: Probabilistic in-band network telemetry,” in Proc. Annu. Conf. ACM Special Interest Group Data Commun. Appl., Technol., Architec., Protocols Comput. Commun., 2020, pp. 662–680.
[58]
S. Tang, D. Li, B. Niu, J. Peng, and Z. Zhu, “Sel-INT: A runtimeprogrammable selective in-band network telemetry system,” IEEE Trans. Netw. Service Manag., vol. 17, no. 2, pp. 708–721, Jun. 2020.
[59]
A. Gholami, S. Kim, Z. Dong, Z. Yao, M. W. Mahoney, and K. Keutzer, “A survey of quantization methods for efficient neural network inference,” 2021, arXiv:2103.13630.
[60]
B. Jacob et al., “Quantization and training of neural networks for efficient integer-arithmetic-only inference,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., 2018, pp. 2704–2713.
[61]
J. Yan et al., “Brain-on-switch: Towards advanced intelligent network data plane via NN-driven traffic analysis at line-speed,” 2024, arXiv:2403.11090.
[62]
M. Ghashami, E. Liberty, J. M. Phillips, and D. P. Woodruff, “Frequent directions: Simple and deterministic matrix sketching,” SIAM J. Comput., vol. 45, no. 5, pp. 1762–1792, 2016.
[63]
K. Razavi, G. Karlos, V. Nigade, M. Mühlhäuser, and L. Wang, “Distributed DNN serving in the network data plane,” in Proc. 5th Int. Workshop P4 Eur., 2022, pp. 67–70.
[64]
D. Koeplinger et al. “Spatial: A language and compiler for application accelerators,” in Proc. 39th ACM SIGPLAN Conf. Program. Lang. Design Implement., 2018, pp. 296–311.
[65]
D. Blalock and J. Guttag, “Multiplying matrices without multiplying,” in Proc. Int. Conf. Mach. Learn., 2021, pp. 992–1004.
[66]
Y. Jeon, B. Park, S. J. Kwon, B. Kim, J. Yun, and D. Lee, “BiQGEMM: Matrix multiplication with lookup table for binary-coding-based quantized DNNs,” in Proc. Int. Conf. High Perform. Comput., Netw., Stor. Anal. (SC20), 2020, pp. 1–14.
[67]
C. Sakr, Y. Kim, and N. Shanbhag, “Analytical guarantees on numerical precision of deep neural networks,” in Proc. Int. Conf. Mach. Learn., 2017, pp. 3007–3016.
[68]
C. Caraiscos and B. Liu, “A roundoff error analysis of the LMS adaptive algorithm,” IEEE Trans. Acoust., Speech, Signal Process., vol. 32, no. 1, pp. 34–41, Feb. 1984.
[69]
J. L. Holt and J.-N. Hwang, “Finite precision error analysis of neural network hardware implementations,” IEEE Trans. Comput., vol. 42, no. 3, pp. 281–290, Mar. 1993.
[70]
I. Sharafaldin, A. H. Lashkari, and A. A. Ghorbani, “Toward generating a new intrusion detection dataset and intrusion traffic characterization,” in Proc. ICISSp, 2018, pp. 108–116.
[71]
G. Draper-Gil, A. H. Lashkari, M. S. Islam Mamun, and A. A. Ghorbani, “Characterization of encrypted and VPN traffic using time-related,” in Proc. 2nd Int. Conf. Inf. Syst. Security Privacy (ICISSP), 2016, pp. 407–414.
[72]
O. Aouedi, K. Piamrat, and B. Parrein, “Ensemble-based deep learning model for network traffic classification,” IEEE Trans. Netw. Service Manag., vol. 19, no. 4, pp. 4124–4135, Dec. 2022.
[73]
M. D. Mauro, G. Galatro, and A. Liotta, “Experimental review of neuralbased approaches for network intrusion management,” IEEE Trans. Netw. Service Manag., vol. 17, no. 4, pp. 2480–2495, Dec. 2020.
[74]
G. Siracusano and R. Bifulco, “In-network neural networks,” 2018, arXiv:1801.05731.
[75]
A. Alowa and T. Fevens, “Towards minimum inter-controller delay time in software defined networking,” Procedia Comput. Sci., vol. 175, pp. 395–402, Aug. 2020.
[76]
J. Xing, K.-F. Hsu, M. Kadosh, A. Lo, Y. Piasetzky, A. Krishnamurthy, and A. Chen, “Runtime programmable switches,” in Proc. 19th USENIX Symp. Netw. Syst. Design Implement. (NSDI 22), 2022, pp. 651–665.
[77]
Y. Feng et al., “Enabling in-situ programmability in network data plane: From architecture to language,” in Proc. 19th USENIX Symp. Netw. Syst. Design Implement. (NSDI 22), 2022, pp. 635–649.
[78]
T. Liang, J. Glossner, L. Wang, S. Shi, and X. Zhang, “Pruning and quantization for deep neural network acceleration: A survey,” Neurocomputing, vol. 461, pp. 370–403, Oct. 2021.
[79]
F. Hauser et al., “A survey on data plane programming with P4: Fundamentals, advances, and applied research,” J. Netw. Comput. Appl., vol. 212, Mar. 2023, Art. no.
Index Terms
- A Machine Learning-Based Toolbox for P4 Programmable Data-Planes
Index terms have been assigned to the content through auto-classification.
Recommendations
Big data, lifelong machine learning and transfer learning
WSDM '13: Proceedings of the sixth ACM international conference on Web search and data miningA major challenge in today's world is the Big Data problem, which manifests itself in Web and Mobile domains as rapidly changing and heterogeneous data streams. A data-mining system must be able to cope with the influx of changing data in a continual ...
Autoware Toolbox: MATLAB/Simulink Benchmark Suite for ROS-based Self-driving Software Platform
RSP '19: Proceedings of the 30th International Workshop on Rapid System Prototyping (RSP'19)This paper describes a MATLAB/Simulink benchmark suite for an open-source self-driving system based on Robot Operating System (ROS). In recent years, self-driving systems have been developed worldwide, and the technology has been making remarkable ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
1932-4537 © 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.
Publisher
IEEE Press
Publication History
Published: 16 May 2024
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 16 Nov 2024
Other Metrics
Citations
View Options
View options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in