DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints

Abdenour Yasser Brahmi; Massinissa Ait Aba; Hadil Bouasker; Badii Jouaber; Hind Castel-Taleb

doi:10.1109/NetSoft64993.2025.11080579

DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints

Abdenour Yasser Brahmi
, Massinissa Ait Aba
, Hadil Bouasker
, Badii Jouaber
, Hind Castel-Taleb

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

Abstract

The Fifth and Sixth Generation (5G/6G) networks aim to support diverse applications with specific QoS and resource needs. Network Slicing has emerged as a key paradigm to meet these demands by creating multiple Virtual Networks (VNs) over shared physical infrastructure. This process, known as Virtual Network Embedding (VNE), maps virtual nodes and links to physical resources. With Kubernetes becoming the dominant orchestration platform, most infrastructures now rely on Kubernetes clusters, which enforce direct pod-to-pod communication, necessitating a direct-link approach to VNE. However, most existing methods focus on path-based link mapping. In this paper, we present DL-ViNE, a Reinforcement Learning(RL)-based algorithm that improves slice acceptance while addressing the specific constraints of Kubernetes-hosted infrastructures.

Original language	English
Title of host publication	Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025
Editors	Pal Varga, Walter Cerroni, Carol Fung, Robert Szabo, Massimo Tornatore
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	448-452
Number of pages	5
ISBN (Electronic)	9798331543457
DOIs	https://doi.org/10.1109/NetSoft64993.2025.11080579
Publication status	Published - 1 Jan 2025
Event	11th IEEE International Conference on Network Softwarization, NetSoft 2025 - Budapest, Hungary Duration: 23 Jun 2025 → 27 Jun 2025

Publication series

Name	Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025

Conference

Conference	11th IEEE International Conference on Network Softwarization, NetSoft 2025
Country/Territory	Hungary
City	Budapest
Period	23/06/25 → 27/06/25

Keywords

5G
Kubernetes
Network Slicing
RL
VNE

Access to Document

10.1109/NetSoft64993.2025.11080579

Cite this

Brahmi, A. Y., Ait Aba, M., Bouasker, H., Jouaber, B., & Castel-Taleb, H. (2025). DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints. In P. Varga, W. Cerroni, C. Fung, R. Szabo, & M. Tornatore (Eds.), Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025 (pp. 448-452). (Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/NetSoft64993.2025.11080579

Brahmi, Abdenour Yasser ; Ait Aba, Massinissa ; Bouasker, Hadil et al. / DL-ViNE : Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints. Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025. editor / Pal Varga ; Walter Cerroni ; Carol Fung ; Robert Szabo ; Massimo Tornatore. Institute of Electrical and Electronics Engineers Inc., 2025. pp. 448-452 (Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025).

@inproceedings{9978a3a05d714adc8b2c00e9749d36f2,

title = "DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints",

abstract = "The Fifth and Sixth Generation (5G/6G) networks aim to support diverse applications with specific QoS and resource needs. Network Slicing has emerged as a key paradigm to meet these demands by creating multiple Virtual Networks (VNs) over shared physical infrastructure. This process, known as Virtual Network Embedding (VNE), maps virtual nodes and links to physical resources. With Kubernetes becoming the dominant orchestration platform, most infrastructures now rely on Kubernetes clusters, which enforce direct pod-to-pod communication, necessitating a direct-link approach to VNE. However, most existing methods focus on path-based link mapping. In this paper, we present DL-ViNE, a Reinforcement Learning(RL)-based algorithm that improves slice acceptance while addressing the specific constraints of Kubernetes-hosted infrastructures.",

keywords = "5G, Kubernetes, Network Slicing, RL, VNE",

author = "Brahmi, \{Abdenour Yasser\} and \{Ait Aba\}, Massinissa and Hadil Bouasker and Badii Jouaber and Hind Castel-Taleb",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 11th IEEE International Conference on Network Softwarization, NetSoft 2025 ; Conference date: 23-06-2025 Through 27-06-2025",

year = "2025",

month = jan,

day = "1",

doi = "10.1109/NetSoft64993.2025.11080579",

language = "English",

series = "Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "448--452",

editor = "Pal Varga and Walter Cerroni and Carol Fung and Robert Szabo and Massimo Tornatore",

booktitle = "Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025",

}

Brahmi, AY, Ait Aba, M, Bouasker, H, Jouaber, B & Castel-Taleb, H 2025, DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints. in P Varga, W Cerroni, C Fung, R Szabo & M Tornatore (eds), Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025. Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025, Institute of Electrical and Electronics Engineers Inc., pp. 448-452, 11th IEEE International Conference on Network Softwarization, NetSoft 2025, Budapest, Hungary, 23/06/25. https://doi.org/10.1109/NetSoft64993.2025.11080579

DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints. / Brahmi, Abdenour Yasser; Ait Aba, Massinissa; Bouasker, Hadil et al.
Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025. ed. / Pal Varga; Walter Cerroni; Carol Fung; Robert Szabo; Massimo Tornatore. Institute of Electrical and Electronics Engineers Inc., 2025. p. 448-452 (Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025).

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

TY - GEN

T1 - DL-ViNE

T2 - 11th IEEE International Conference on Network Softwarization, NetSoft 2025

AU - Brahmi, Abdenour Yasser

AU - Ait Aba, Massinissa

AU - Bouasker, Hadil

AU - Jouaber, Badii

AU - Castel-Taleb, Hind

PY - 2025/1/1

Y1 - 2025/1/1

N2 - The Fifth and Sixth Generation (5G/6G) networks aim to support diverse applications with specific QoS and resource needs. Network Slicing has emerged as a key paradigm to meet these demands by creating multiple Virtual Networks (VNs) over shared physical infrastructure. This process, known as Virtual Network Embedding (VNE), maps virtual nodes and links to physical resources. With Kubernetes becoming the dominant orchestration platform, most infrastructures now rely on Kubernetes clusters, which enforce direct pod-to-pod communication, necessitating a direct-link approach to VNE. However, most existing methods focus on path-based link mapping. In this paper, we present DL-ViNE, a Reinforcement Learning(RL)-based algorithm that improves slice acceptance while addressing the specific constraints of Kubernetes-hosted infrastructures.

AB - The Fifth and Sixth Generation (5G/6G) networks aim to support diverse applications with specific QoS and resource needs. Network Slicing has emerged as a key paradigm to meet these demands by creating multiple Virtual Networks (VNs) over shared physical infrastructure. This process, known as Virtual Network Embedding (VNE), maps virtual nodes and links to physical resources. With Kubernetes becoming the dominant orchestration platform, most infrastructures now rely on Kubernetes clusters, which enforce direct pod-to-pod communication, necessitating a direct-link approach to VNE. However, most existing methods focus on path-based link mapping. In this paper, we present DL-ViNE, a Reinforcement Learning(RL)-based algorithm that improves slice acceptance while addressing the specific constraints of Kubernetes-hosted infrastructures.

KW - 5G

KW - Kubernetes

KW - Network Slicing

KW - RL

KW - VNE

UR - https://www.scopus.com/pages/publications/105012574834

U2 - 10.1109/NetSoft64993.2025.11080579

DO - 10.1109/NetSoft64993.2025.11080579

M3 - Conference contribution

AN - SCOPUS:105012574834

T3 - Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025

SP - 448

EP - 452

BT - Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025

A2 - Varga, Pal

A2 - Cerroni, Walter

A2 - Fung, Carol

A2 - Szabo, Robert

A2 - Tornatore, Massimo

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 23 June 2025 through 27 June 2025

ER -

Brahmi AY, Ait Aba M, Bouasker H, Jouaber B, Castel-Taleb H. DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints. In Varga P, Cerroni W, Fung C, Szabo R, Tornatore M, editors, Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025. Institute of Electrical and Electronics Engineers Inc. 2025. p. 448-452. (Proceedings of the 11th IEEE International Conference on Network Softwarization, NetSoft 2025). doi: 10.1109/NetSoft64993.2025.11080579

DL-ViNE: Reinforcement Learning Algorithm for Efficient Virtual Network Embedding Under Direct-Link Constraints

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this