Configurable deployment descriptor model in NFV - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Network and Systems Management Année : 2020

Configurable deployment descriptor model in NFV

Résumé

The deployment of network services and functions in Network Function Virtualization (NFV) platforms is performed using deployment descriptors. The deployment descriptors describe the topology of the network service and the operation behavior of the network functions along with their resource requirements. They are huge file written manually by service providers devoid of any formal strategy besides experience. Designing those descriptors from scratch is undoubtedly a tedious and complex task given the dynamicity at which the NFV is constantly evolving. Additionally, relying solely on the service provider knowledge is not only error-prone, but also challengeable. Motivated by the design by reuse paradigm, we present in this paper a configurable descriptor model that allows to formalize best practices in designing descriptor models for a specific domain. A configurable descriptor model can be seen as a generic solution that needs to be configured by a service provider to suit the specific requirements of a deployed service. In this way, configurable descriptor models can ease and accelerate the modeling of deployment descriptors by providing an off-the-shelf-solution that can be easily configured in a “plug and play” fashion. Our prototypical implementation demonstrates the feasibility of our approach and the results of our experiments highlight the effectiveness of our proposed model.

Domaines

Informatique
Fichier non déposé

Dates et versions

hal-03122714 , version 1 (27-01-2021)

Identifiants

Citer

Wassim Sellil Atoui, Nour Assy, Walid Gaaloul, Imen Grida Ben Yahia. Configurable deployment descriptor model in NFV. Journal of Network and Systems Management, 2020, 28 (3), pp.693-718. ⟨10.1007/s10922-020-09531-2⟩. ⟨hal-03122714⟩
32 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More