Scholarlink

<p>The advent of network function virtualization (NFV) has transformed the traditional network infrastructure by enabling the virtualization of network functions, i.e., virtual network functions (VNFs). VNF placement is a challenging problem in cloud-based networks, particularly under service function chain (SFC) constraints, where the ordering and reuse of functions across multiple chains must be considered. The traditional optimization approaches often struggle with the complexity and dynamics of large-scale 5G/6G networks. This survey provides a comprehensive review of the recent advances achieved in VNF placement using machine learning (ML) techniques and explores the integration of the analytic hierarchy process (AHP) into multicriteria evaluation and prioritization schemes. This work summarizes the key methods and parameters; highlights the potential of hybrid ML-AHP approaches; and identifies open research directions for optimizing the performance, resource utilization rates, and quality of service attained in virtualized network environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multicriteria VNF placement and chaining with machine learning: a survey and new opportunities for the AHP in decision-making scenarios

  • 期刊论文

  • 发表日期：

    2026年2月16日

查看全文
  • Zahraa Abed Aljasim Muhisn,
  • Yassine Boujelben,
  • Faouzi Zarai

摘要

The advent of network function virtualization (NFV) has transformed the traditional network infrastructure by enabling the virtualization of network functions, i.e., virtual network functions (VNFs). VNF placement is a challenging problem in cloud-based networks, particularly under service function chain (SFC) constraints, where the ordering and reuse of functions across multiple chains must be considered. The traditional optimization approaches often struggle with the complexity and dynamics of large-scale 5G/6G networks. This survey provides a comprehensive review of the recent advances achieved in VNF placement using machine learning (ML) techniques and explores the integration of the analytic hierarchy process (AHP) into multicriteria evaluation and prioritization schemes. This work summarizes the key methods and parameters; highlights the potential of hybrid ML-AHP approaches; and identifies open research directions for optimizing the performance, resource utilization rates, and quality of service attained in virtualized network environments.