Distributed DNN serving in the network data plane

K. Razavi; G. Karlos; V. Nigade; M. Mühlhäuser; L. Wang

doi:10.1145/3565475.3569079

Distributed DNN serving in the network data plane

K. Razavi
, G. Karlos
, V. Nigade
, M. Mühlhäuser
, L. Wang

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

273 Downloads (Pure)

Abstract

Programmable networks have received tremendous attention recently. Apart from exciting network innovations, in-network computing has been explored as a means to accelerate a variety of distributed systems concerns, by leveraging programmable network devices. In this paper, we extend in-network computing to an important class of applications called deep neural network (DNN) serving. In particular, we propose to run DNN inferences in the network data plane in a distributed fashion and make our programmable network a powerful accelerator for DNN serving. We demonstrate the feasibility of this idea through a case study with a real-world DNN on a typical data center network architecture.

Original language	English
Title of host publication	EuroP4 '22
Subtitle of host publication	Proceedings of the 5th International Workshop on P4 in Europe
Publisher	Association for Computing Machinery, Inc
Pages	67-70
Number of pages	4
ISBN (Electronic)	9781450399357
DOIs	https://doi.org/10.1145/3565475.3569079
Publication status	Published - Dec 2022
Event	5th International Workshop on P4 in Europe, EuroP4 2022, co-located with ACM CoNEXT 2022 - Rome, Italy Duration: 9 Dec 2022 → …

Conference

Conference	5th International Workshop on P4 in Europe, EuroP4 2022, co-located with ACM CoNEXT 2022
Country/Territory	Italy
City	Rome
Period	9/12/22 → …

Funding

Funders	Funder number
Open Competition Domain Science XS
Deutsche Forschungsgemeinschaft
Google Research
???publication-publication-funding-organisation-not-added???	12611

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Access to Document

10.1145/3565475.3569079

Distributed DNN serving in the network data planeFinal published version, 528 KBLicence: Article 25fa Dutch Copyright Act

Persistent URL (handle)

Persistent link

Cite this

@inproceedings{f2dd1703f1e2451d964eb13685ba9de8,

title = "Distributed DNN serving in the network data plane",

abstract = "Programmable networks have received tremendous attention recently. Apart from exciting network innovations, in-network computing has been explored as a means to accelerate a variety of distributed systems concerns, by leveraging programmable network devices. In this paper, we extend in-network computing to an important class of applications called deep neural network (DNN) serving. In particular, we propose to run DNN inferences in the network data plane in a distributed fashion and make our programmable network a powerful accelerator for DNN serving. We demonstrate the feasibility of this idea through a case study with a real-world DNN on a typical data center network architecture.",

author = "K. Razavi and G. Karlos and V. Nigade and M. M{\"u}hlh{\"a}user and L. Wang",

year = "2022",

month = dec,

doi = "10.1145/3565475.3569079",

language = "English",

pages = "67--70",

booktitle = "EuroP4 '22",

publisher = "Association for Computing Machinery, Inc",

note = "5th International Workshop on P4 in Europe, EuroP4 2022, co-located with ACM CoNEXT 2022 ; Conference date: 09-12-2022",

}

Razavi, K, Karlos, G, Nigade, V, Mühlhäuser, M & Wang, L 2022, Distributed DNN serving in the network data plane. in EuroP4 '22: Proceedings of the 5th International Workshop on P4 in Europe. Association for Computing Machinery, Inc, pp. 67-70, 5th International Workshop on P4 in Europe, EuroP4 2022, co-located with ACM CoNEXT 2022, Rome, Italy, 9/12/22. https://doi.org/10.1145/3565475.3569079

TY - GEN

T1 - Distributed DNN serving in the network data plane

AU - Razavi, K.

AU - Karlos, G.

AU - Nigade, V.

AU - Mühlhäuser, M.

AU - Wang, L.

PY - 2022/12

Y1 - 2022/12

N2 - Programmable networks have received tremendous attention recently. Apart from exciting network innovations, in-network computing has been explored as a means to accelerate a variety of distributed systems concerns, by leveraging programmable network devices. In this paper, we extend in-network computing to an important class of applications called deep neural network (DNN) serving. In particular, we propose to run DNN inferences in the network data plane in a distributed fashion and make our programmable network a powerful accelerator for DNN serving. We demonstrate the feasibility of this idea through a case study with a real-world DNN on a typical data center network architecture.

AB - Programmable networks have received tremendous attention recently. Apart from exciting network innovations, in-network computing has been explored as a means to accelerate a variety of distributed systems concerns, by leveraging programmable network devices. In this paper, we extend in-network computing to an important class of applications called deep neural network (DNN) serving. In particular, we propose to run DNN inferences in the network data plane in a distributed fashion and make our programmable network a powerful accelerator for DNN serving. We demonstrate the feasibility of this idea through a case study with a real-world DNN on a typical data center network architecture.

UR - https://www.scopus.com/pages/publications/85145596391

UR - https://www.scopus.com/pages/publications/85145596391#tab=citedBy

U2 - 10.1145/3565475.3569079

DO - 10.1145/3565475.3569079

M3 - Conference contribution

SP - 67

EP - 70

BT - EuroP4 '22

PB - Association for Computing Machinery, Inc

T2 - 5th International Workshop on P4 in Europe, EuroP4 2022, co-located with ACM CoNEXT 2022

Y2 - 9 December 2022

ER -

Distributed DNN serving in the network data plane

Abstract

Conference

Funding

UN SDGs

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this