Towards a generic deep learning pipeline for traffic measurements

Raphaël Azorin; Massimo Gallo; Alessandro Finamore; Maurizio Filippone; Pietro Michiardi; Dario Rossi

doi:10.1145/3488658.3493785

Towards a generic deep learning pipeline for traffic measurements

Raphaël Azorin, Massimo Gallo, Alessandro Finamore, Maurizio Filippone, Pietro Michiardi, Dario Rossi

Statistics

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

Abstract

Traffic measurements are key for network management as testified by the rich literature from both academia and industry. At their foundation, measurements rely on transformation functions f(x) = y, mapping input traffic data x to an output performance metric y. Yet, common practices adopt a bottom-up design (i.e., metric-based) which leads to (i) invest a lot of efforts into (re)discovering how to perform such mapping and (ii) create specialized solutions. For instance, sketches are a compact way to extract traffic properties (heavy-hitters, super-spreaders, etc.) but require analytical modeling to offer correctness guarantees and careful engineering to enable in-device deployment and network-wide measurements.

Original language	English (US)
Title of host publication	CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies
Publisher	Association for Computing Machinery, Inc
Pages	5-6
Number of pages	2
ISBN (Electronic)	9781450391337
DOIs	https://doi.org/10.1145/3488658.3493785
State	Published - Dec 7 2021
Event	2nd ACM CoNEXT Student Workshop, CoNEXT-SW 2021, co-located with the 17th International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021 - Virtual, Online, Germany Duration: Dec 7 2021 → …

Publication series

Name	CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies

Conference

Conference	2nd ACM CoNEXT Student Workshop, CoNEXT-SW 2021, co-located with the 17th International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021
Country/Territory	Germany
City	Virtual, Online
Period	12/7/21 → …

Bibliographical note

Publisher Copyright:
© 2021 ACM.

Keywords

deep learning
network measurements
representation learning

ASJC Scopus subject areas

Hardware and Architecture
Computer Science Applications
Computer Networks and Communications

Access to Document

10.1145/3488658.3493785

Cite this

Azorin, R., Gallo, M., Finamore, A., Filippone, M., Michiardi, P., & Rossi, D. (2021). Towards a generic deep learning pipeline for traffic measurements. In CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies (pp. 5-6). (CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies). Association for Computing Machinery, Inc. https://doi.org/10.1145/3488658.3493785

Azorin, Raphaël ; Gallo, Massimo ; Finamore, Alessandro et al. / Towards a generic deep learning pipeline for traffic measurements. CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies. Association for Computing Machinery, Inc, 2021. pp. 5-6 (CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies).

@inproceedings{249e8de1d1b646a985feddde3abf458b,

title = "Towards a generic deep learning pipeline for traffic measurements",

abstract = "Traffic measurements are key for network management as testified by the rich literature from both academia and industry. At their foundation, measurements rely on transformation functions f(x) = y, mapping input traffic data x to an output performance metric y. Yet, common practices adopt a bottom-up design (i.e., metric-based) which leads to (i) invest a lot of efforts into (re)discovering how to perform such mapping and (ii) create specialized solutions. For instance, sketches are a compact way to extract traffic properties (heavy-hitters, super-spreaders, etc.) but require analytical modeling to offer correctness guarantees and careful engineering to enable in-device deployment and network-wide measurements.",

keywords = "deep learning, network measurements, representation learning",

author = "Rapha{\"e}l Azorin and Massimo Gallo and Alessandro Finamore and Maurizio Filippone and Pietro Michiardi and Dario Rossi",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 2nd ACM CoNEXT Student Workshop, CoNEXT-SW 2021, co-located with the 17th International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021 ; Conference date: 07-12-2021",

year = "2021",

month = dec,

day = "7",

doi = "10.1145/3488658.3493785",

language = "English (US)",

series = "CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies",

publisher = "Association for Computing Machinery, Inc",

pages = "5--6",

booktitle = "CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies",

}

Azorin, R, Gallo, M, Finamore, A, Filippone, M, Michiardi, P & Rossi, D 2021, Towards a generic deep learning pipeline for traffic measurements. in CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies. CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies, Association for Computing Machinery, Inc, pp. 5-6, 2nd ACM CoNEXT Student Workshop, CoNEXT-SW 2021, co-located with the 17th International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021, Virtual, Online, Germany, 12/7/21. https://doi.org/10.1145/3488658.3493785

Towards a generic deep learning pipeline for traffic measurements. / Azorin, Raphaël; Gallo, Massimo; Finamore, Alessandro et al.
CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies. Association for Computing Machinery, Inc, 2021. p. 5-6 (CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies).

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

TY - GEN

T1 - Towards a generic deep learning pipeline for traffic measurements

AU - Azorin, Raphaël

AU - Gallo, Massimo

AU - Finamore, Alessandro

AU - Filippone, Maurizio

AU - Michiardi, Pietro

AU - Rossi, Dario

PY - 2021/12/7

Y1 - 2021/12/7

N2 - Traffic measurements are key for network management as testified by the rich literature from both academia and industry. At their foundation, measurements rely on transformation functions f(x) = y, mapping input traffic data x to an output performance metric y. Yet, common practices adopt a bottom-up design (i.e., metric-based) which leads to (i) invest a lot of efforts into (re)discovering how to perform such mapping and (ii) create specialized solutions. For instance, sketches are a compact way to extract traffic properties (heavy-hitters, super-spreaders, etc.) but require analytical modeling to offer correctness guarantees and careful engineering to enable in-device deployment and network-wide measurements.

AB - Traffic measurements are key for network management as testified by the rich literature from both academia and industry. At their foundation, measurements rely on transformation functions f(x) = y, mapping input traffic data x to an output performance metric y. Yet, common practices adopt a bottom-up design (i.e., metric-based) which leads to (i) invest a lot of efforts into (re)discovering how to perform such mapping and (ii) create specialized solutions. For instance, sketches are a compact way to extract traffic properties (heavy-hitters, super-spreaders, etc.) but require analytical modeling to offer correctness guarantees and careful engineering to enable in-device deployment and network-wide measurements.

KW - deep learning

KW - network measurements

KW - representation learning

UR - http://www.scopus.com/inward/record.url?scp=85121581146&partnerID=8YFLogxK

U2 - 10.1145/3488658.3493785

DO - 10.1145/3488658.3493785

M3 - Conference contribution

AN - SCOPUS:85121581146

T3 - CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies

SP - 5

EP - 6

BT - CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies

PB - Association for Computing Machinery, Inc

T2 - 2nd ACM CoNEXT Student Workshop, CoNEXT-SW 2021, co-located with the 17th International Conference on emerging Networking EXperiments and Technologies, CoNEXT 2021

Y2 - 7 December 2021

ER -

Azorin R, Gallo M, Finamore A, Filippone M, Michiardi P, Rossi D. Towards a generic deep learning pipeline for traffic measurements. In CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies. Association for Computing Machinery, Inc. 2021. p. 5-6. (CoNEXT-SW 2021 - Proceedings of the 2021 CoNEXT Student Workshop 2021 - Part of CoNEXT 2021 the 17th International Conference on emerging Networking EXperiments and Technologies). doi: 10.1145/3488658.3493785

Towards a generic deep learning pipeline for traffic measurements

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this