A slice-based decentralized NFV framework for an end-to-end QoS-based dynamic resource allocation
- PDF / 3,289,079 Bytes
- 19 Pages / 595.276 x 790.866 pts Page_size
- 93 Downloads / 191 Views
ORIGINAL RESEARCH
A slice‑based decentralized NFV framework for an end‑to‑end QoS‑based dynamic resource allocation Inès Raïssa Djouela Kamgang1 · Ghayet El Mouna Zhioua1 · Nabil Tabbane1 Received: 1 July 2019 / Accepted: 7 January 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Network function virtualization concept has recently merged to solve network operators and service provider’s problems related to the non-flexibility of the traditional network and the increase of capital and operational expenditures (CAPEX and OPEX). ETSI has standardized an architectural framework to serve as a springboard for reflection to the application and the setup of the concept. However, that architecture presents many shortcomings and several challenges that have to be addressed. In this paper, we propose a new original framework that aims at satisfying the subscriber’s Service Level Agreement and optimizes Telecom Service Provider fees under the scope of pay as you go concept. Based on the proposed framework, the end to end service execution process is presented. We afterward address the dynamic scheduling problem to minimize service execution time in the network by meant of the makespan. The dynamic scheduling problem is modeled as a mixed-integer linear programming one, and an Event-Driven Upward Ranking-Based Dynamic Scheduling Heuristic (EDURBaDySH) is proposed to solve large instances of the problem. Simulation results show that our approach performs better than Round Robin, Min Min and HEFT approaches in terms of makespan for different workflow configurations, considering a small, middle and large number of VNFs. Keywords Network function virtualization · Resource allocation · Chaining · Mapping∖dynamic scheduling · Slices · Pay as you go
1 Introduction Telecommunication networks have traditionally been designed to support heterogeneous types of services, e.g., voice, data, etc. With that configuration, these services are obtained by statically combining network functions (NFs) deployed in hardware middleboxes following a strictly ordered chain. That traditional configuration is, however, becoming limited due to the lack of flexibility and the high operator’s capital and operational expenditures (CAPEX and OPEX).
* Inès Raïssa Djouela Kamgang [email protected] Ghayet El Mouna Zhioua [email protected] Nabil Tabbane [email protected] 1
Higher School of Communication of Tunis, Mediatron Laboratory, University of Carthage, Tunis, Tunisia
In fact, the traditional configuration remains very expensive for operators, since the hardware middleboxes are proprietary and the built-in features are unique, requiring renewal at the slightest technological change. Furthermore, the deployment of the plethora of 5G new services such as autonomous car driving, connected homes, etc. is infeasible with the traditional network as they are greedy in resource consumption and require high quality of services (QoS) at low costs. Network function virtualization (NFV) is a new network architecture p
Data Loading...
