SFC Provisioning over NFV Enabled Clouds
Autor: | Marouen Mechtri, Oussama Soualah, Chaima Ghribi, Djamal Zeghlache |
---|---|
Přispěvatelé: | Département Réseaux et Services Multimédia Mobiles (RS2M), Institut Mines-Télécom [Paris] (IMT)-Télécom SudParis (TSP), Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux (SAMOVAR), Centre National de la Recherche Scientifique (CNRS), Réseaux, Systèmes, Services, Sécurité (R3S-SAMOVAR), Institut Mines-Télécom [Paris] (IMT)-Télécom SudParis (TSP)-Institut Mines-Télécom [Paris] (IMT)-Télécom SudParis (TSP) |
Rok vydání: | 2017 |
Předmět: |
Recursive approach
Optimization problem Computer science Distributed computing Cloud computing 02 engineering and technology NFV [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI] Bandwidth Hardware Mathematical model Robustness (computer science) 0202 electrical engineering electronic engineering information engineering Heuristic algorithms Revenue Service function chaining Substrates business.industry Testbed Computational modeling 020206 networking & telecommunications Provisioning Chaining 020201 artificial intelligence & image processing business |
Zdroj: | CLOUD Proceedings CLOUD 2017 : IEEE 10th International Conference on Cloud Computing CLOUD 2017 : IEEE 10th International Conference on Cloud Computing CLOUD 2017 : IEEE 10th International Conference on Cloud Computing, Jun 2017, Honolulu, United States. pp.423-430, ⟨10.1109/CLOUD.2017.61⟩ |
Popis: | International audience; This paper deals with virtualised Network Function Placement and chaining over cloud and NFV environments. We propose a polynomial-time algorithm for joint VNF chain placement that is based on a recursive optimization approach. The key idea is to benefit from the robustness of the recursion paradigm that was used to solve many optimization problems. We demonstrate the potential of our solution by comparing it to previous work. We conduct performance evaluation first through extensive simulations and then using a real cloud testbed. The obtained results show that our proposed algorithm achieves better performance in terms of resource usage, execution time, acceptance rate and revenue of the cloud provider |
Databáze: | OpenAIRE |
Externí odkaz: |