Traffic Refinery: Cost-Aware Data Representation for Machine Learning on Network Traffic

被引:9
|
作者
Bronzino, Francesco [1 ]
Schmitt, Paul [2 ]
Ayoubi, Sara [3 ]
Kim, Hyojoon [4 ]
Teixeira, Renata [5 ]
Feamster, Nick [6 ]
机构
[1] Univ Savoie Mt Blanc, LISTIC, Annecy Le Vieux, France
[2] USC Informat Sci Inst, Los Angeles, CA USA
[3] Nokia Bell Labs, Paris Saclay, France
[4] Princeton Univ, Princeton, NJ 08544 USA
[5] Inria, Paris, France
[6] Univ Chicago, Chicago, IL 60637 USA
关键词
network systems; network traffic; QoS inference; malware detection;
D O I
10.1145/3491052
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Network management often relies on machine learning to make predictions about performance and security from network traffic. Often, the representation of the traffic is as important as the choice of the model. The features that the model relies on, and the representation of those features, ultimately determine model accuracy, as well as where and whether the model can be deployed in practice. Thus, the design and evaluation of these models ultimately requires understanding not only model accuracy but also the systems costs associated with deploying the model in an operational network. Towards this goal, this paper develops a new framework and system that enables a joint evaluation of both the conventional notions of machine learning performance (e.g., model accuracy) and the systems-level costs of different representations of network traffic. We highlight these two dimensions for two practical network management tasks, video streaming quality inference and malware detection, to demonstrate the importance of exploring different representations to find the appropriate operating point. We demonstrate the benefit of exploring a range of representations of network traffic and present Traffic Refinery, a proof-of-concept implementation that both monitors network traffic at 10 Gbps and transforms traffic in real time to produce a variety of feature representations for machine learning. Traffic Refinery both highlights this design space and makes it possible to explore different representations for learning, balancing systems costs related to feature extraction and model training against model accuracy.
引用
收藏
页数:24
相关论文
共 50 条
  • [41] Cost-based virtual machine scheduling resilient to network traffic anomalies for Data as a Service
    Alves de Oliveira, Ana Cristina
    Spohn, Marco Aurelio
    REVISTA BRASILEIRA DE COMPUTACAO APLICADA, 2020, 12 (03): : 85 - 96
  • [42] Data representation model for in-depth analysis of network traffic
    Get'man, A. I.
    Ivannikov, V. P.
    Markin, Yu. V.
    Padaryan, V. A.
    Tikhonov, A. Yu.
    PROGRAMMING AND COMPUTER SOFTWARE, 2016, 42 (05) : 316 - 323
  • [43] Data representation model for in-depth analysis of network traffic
    I. Get’man
    V. P. Ivannikov
    Yu. V. Markin
    V. A. Padaryan
    A. Yu. Tikhonov
    Programming and Computer Software, 2016, 42 : 316 - 323
  • [44] Fast Context Adaptation in Cost-Aware Continual Learning
    Lahmer, Seyyidahmed
    Mason, Federico
    Chiariotti, Federico
    Zanella, Andrea
    IEEE Transactions on Machine Learning in Communications and Networking, 2024, 2 : 479 - 494
  • [45] Experimental Design for Cost-Aware Learning of Causal Graphs
    Lindgren, Erik M.
    Kocaoglu, Murat
    Dimakis, Alexandros G.
    Vishwanath, Sriram
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [46] Cost-Aware Active Learning for Feasible Region Identification
    Nikova, Ioana
    Dhaene, Tom
    Couckuyt, Ivo
    PROCEEDINGS OF THE 2023 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE COMPANION, GECCO 2023 COMPANION, 2023, : 2286 - 2288
  • [47] Cost-Aware Learning for Improved Identifiability with Multiple Experiments
    Guo, Longyun
    Honorio, Jean
    Morgan, John
    2019 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2019, : 1802 - 1806
  • [48] Cost-Aware Learning and Optimization for Opportunistic Spectrum Access
    Gan, Chao
    Zhou, Ruida
    Yang, Jing
    Shen, Cong
    2020 INFORMATION THEORY AND APPLICATIONS WORKSHOP (ITA), 2020,
  • [49] Cost-Aware Learning and Optimization for Opportunistic Spectrum Access
    Gan, Chao
    Zhou, Ruida
    Yang, Jing
    Shen, Cong
    IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING, 2019, 5 (01) : 15 - 27
  • [50] Reinforcement Learning for Cost-Aware Markov Decision Processes
    Suttle, Wesley A.
    Zhang, Kaiqing
    Yang, Zhuoran
    Kraemer, David N.
    Liu, Ji
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 139, 2021, 139