A multi-objective non-dominated sorting genetic algorithm for VNF chains placement

Selma Khebbache 1, 2 Makhlouf Hadji 3 Zeghlache Djamal 2, 1, 4
1 RS2M - Département Réseaux et Services Multimédia Mobiles
2 SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
3 IRT SystemX - IRT SystemX
4 R3S-SAMOVAR - Réseaux, Systèmes, Services, Sécurité
SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
Abstract : We propose a meta-heuristic based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to address the NP-Hard service function chain placement problem. This work considers the minimization of the mapping cost and of the physical links utilization for virtualized network functions (VNF) chaining. The proposed NSGA-II based algorithm finds a Pareto front to select solutions that meet the multiple objectives and performance tradeoffs of providers. Simulation results and comparison with a multi-stage algorithm and a matrix based heuristic from the literature, highlight the efficiency and usefulness of the proposed NSGA-II-based approach
Keywords : NFV Optimization NSGA-II Forwarding graph Chains placement
Type de document :
Communication dans un congrès
CCNC 2018 : IEEE Consumer Communications & Networking Conference, Jan 2018, Las Vegas, United States. 2018, 〈10.1109/CCNC.2018.8319250〉
Liste complète des métadonnées
https://hal.archives-ouvertes.fr/hal-01613473
Contributeur : Makhlouf Hadji <>
Soumis le : lundi 9 octobre 2017 - 16:16:36
Dernière modification le : mardi 22 janvier 2019 - 14:32:09

Identifiants

Collections

IRT-SYSTEMX | INSTITUT-TELECOM | TELECOM-SUDPARIS | UNIV-PARIS-SACLAY

Citation

Selma Khebbache, Makhlouf Hadji, Zeghlache Djamal. A multi-objective non-dominated sorting genetic algorithm for VNF chains placement. CCNC 2018 : IEEE Consumer Communications & Networking Conference, Jan 2018, Las Vegas, United States. 2018, 〈10.1109/CCNC.2018.8319250〉. 〈hal-01613473〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

269

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD