Loading…

Placement Optimization of Virtual Network Functions in a Cloud Computing Environment

The use of Network Function Virtualization is constantly increasing in Cloud environments, especially for next-generation networks such as 5G. In this context, the definition of a deployment scheme defining for each Virtual Network Function (VNF) the appropriate server in order to meet the quality o...

Full description

Saved in:
Bibliographic Details
Published in:Journal of network and systems management 2024-04, Vol.32 (2), p.39, Article 39
Main Authors: Said, Imad Eddine, Sayad, Lamri, Aissani, Djamil
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The use of Network Function Virtualization is constantly increasing in Cloud environments, especially for next-generation networks such as 5G. In this context, the definition of a deployment scheme defining for each Virtual Network Function (VNF) the appropriate server in order to meet the quality of service requirements. This problem is known in the literature as virtual fetwork function placement. However, proper deployment of VNFs on servers can minimize the number of servers used, but may increase service latency. In this article, we propose a multi-objective integer linear programming model to solve the problem of network function placement. The objective is to find the best compromise between minimizing end-to-end total latency for users and reducing the number of servers used, while ensuring that the maximum number of VNFs is connected in the network. Our proposal to solve the NP-hard problem involves developing an algorithm based on the Particle Swarm Optimization metaheuristic to obtain a polynomial time resolution. By performing tests on a simple VNF deployment problem, we validated the relevance of our optimization model and demonstrated the effectiveness of our algorithm. The results obtained showed that our method provides feasible solutions very close to the exact optimal solutions.
ISSN:1064-7570
1573-7705
DOI:10.1007/s10922-024-09812-0