Multi-objective VNF Placement Optimization with NSGA-III

Mahmoud Bekhit, Ahmed Fathalla, Esraa Eldesouky, Ahmad Salah

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

With the rising acceptance of virtual network functions (VNFs) as a replacement for traditional network functions, the optimal placement of VNFs has become a crucial task for ensuring constant performance within constrained resources. This research investigates the challenge of mapping and scheduling VNFs as a multi-objective optimization problem. The state-of-the-art methods have focused on optimizing the mapping or scheduling of VNFs while considering one or two objectives continuously. In this study, we aim to optimize two objectives: minimizing the link capacity cost and maximizing the average resource utilization of each virtual machine. To address this problem, an adaptation of the NSGA-III optimization algorithm is proposed and thoroughly evaluated through a set of experiments. In terms of objective values, the results outlined that the proposed method outperformed the other methods of comparison by a significant margin.

Original languageEnglish
Title of host publicationProceedings of the 2023 International Conference on Advances in Computing Research (ACR’23)
EditorsKevin Daimi, Abeer Al Sadoon
PublisherSpringer Science and Business Media Deutschland GmbH
Pages481-493
Number of pages13
ISBN (Print)9783031337420
DOIs
StatePublished - 2023
Event1st International Conference on Advances in Computing Research, ACR’23 - Orlando, United States
Duration: 8 May 202310 May 2023

Publication series

NameLecture Notes in Networks and Systems
Volume700 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference1st International Conference on Advances in Computing Research, ACR’23
Country/TerritoryUnited States
CityOrlando
Period8/05/2310/05/23

Keywords

  • Multi-objective optimization
  • NSGA-III
  • Optimization algorithms
  • Resource allocation
  • VNF scheduling and mapping

Fingerprint

Dive into the research topics of 'Multi-objective VNF Placement Optimization with NSGA-III'. Together they form a unique fingerprint.

Cite this