Comparative analysis of load balancing methods based on SDN/NFV
Uloženo v:
| Název: | Comparative analysis of load balancing methods based on SDN/NFV |
|---|---|
| Autoři: | O. Berestovenko |
| Zdroj: | Ìнформаційні технології та компʼютерна інженерія, Vol 22, Iss 1, Pp 124-133 (2025) |
| Informace o vydavateli: | Vinnytsia National Technical University, 2025. |
| Rok vydání: | 2025 |
| Sbírka: | LCC:Computer engineering. Computer hardware |
| Témata: | network functions virtualisation, intelligent algorithms, resource optimisation, hybrid structures, energy efficiency, Computer engineering. Computer hardware, TK7885-7895 |
| Popis: | The purpose of the study was to determine the advantages and disadvantages of using load balancing methods in networks based on Software-Defined Networking and Network Functions Virtualisation technologies. Particular attention was paid to comparing the effectiveness of different approaches to balancing, including centralised and distributed methods, and the use of intelligent algorithms for load forecasting. The analysis helped to identify the advantages and disadvantages of each method, and their ability to adapt to changing network traffic conditions, considering such parameters as bandwidth, latency, packet loss, and energy efficiency. The paper discussed methods of load balancing in networks based on Software-Defined Networking and Network Functions Virtualisation technologies, which are important for ensuring the efficiency, scalability, and adaptability of modern networks. The key challenges faced by these technologies are described, such as the dynamism and unpredictability of traffic, resource optimisation, energy efficiency, and the integration of intelligent algorithms for load forecasting and energy consumption reduction. The study presents a comparison of various load balancing methods, including centralised and distributed traffic management, and the use of virtual balancers and adaptive traffic redirection algorithms. Particular attention was paid to analysing the impact of these methods on throughput, latency, packet loss, and energy efficiency under different traffic conditions. The role of machine learning in optimising load balancing processes, and the possibilities of integrating Software-Defined Networking and Network Functions Virtualisation into hybrid networks were considered. According to the results of the study, the use of balancing methods based on Software-Defined Networking/Network Functions Virtualisation can significantly improve network efficiency, reduce latency and increase throughput, while reducing energy consumption under high loads. Key results for Ukraine, where the integration of Software-Defined Networking/Network Functions Virtualisation into the telecommunications infrastructure can become the basis for improving the quality of services, optimising costs, and ensuring a high level of security in the context of digital transformation and infrastructure modernisation, are derived |
| Druh dokumentu: | article |
| Popis souboru: | electronic resource |
| Jazyk: | English Ukrainian |
| ISSN: | 1999-9941 2078-6387 |
| Relation: | https://itce.vn.ua/journals/t-22-1-2025/porivnyalny-analiz-metodiv-balansuvannya-navantazhennya-na-osnovi-sdn-nfv; https://doaj.org/toc/1999-9941; https://doaj.org/toc/2078-6387 |
| DOI: | 10.63341/vitce/1.2025.124 |
| Přístupová URL adresa: | https://doaj.org/article/a7efc89fe8c749d695e72d647570ffef |
| Přístupové číslo: | edsdoj.7efc89fe8c749d695e72d647570ffef |
| Databáze: | Directory of Open Access Journals |
Nájsť tento článok vo Web of Science | Abstrakt: | The purpose of the study was to determine the advantages and disadvantages of using load balancing methods in networks based on Software-Defined Networking and Network Functions Virtualisation technologies. Particular attention was paid to comparing the effectiveness of different approaches to balancing, including centralised and distributed methods, and the use of intelligent algorithms for load forecasting. The analysis helped to identify the advantages and disadvantages of each method, and their ability to adapt to changing network traffic conditions, considering such parameters as bandwidth, latency, packet loss, and energy efficiency. The paper discussed methods of load balancing in networks based on Software-Defined Networking and Network Functions Virtualisation technologies, which are important for ensuring the efficiency, scalability, and adaptability of modern networks. The key challenges faced by these technologies are described, such as the dynamism and unpredictability of traffic, resource optimisation, energy efficiency, and the integration of intelligent algorithms for load forecasting and energy consumption reduction. The study presents a comparison of various load balancing methods, including centralised and distributed traffic management, and the use of virtual balancers and adaptive traffic redirection algorithms. Particular attention was paid to analysing the impact of these methods on throughput, latency, packet loss, and energy efficiency under different traffic conditions. The role of machine learning in optimising load balancing processes, and the possibilities of integrating Software-Defined Networking and Network Functions Virtualisation into hybrid networks were considered. According to the results of the study, the use of balancing methods based on Software-Defined Networking/Network Functions Virtualisation can significantly improve network efficiency, reduce latency and increase throughput, while reducing energy consumption under high loads. Key results for Ukraine, where the integration of Software-Defined Networking/Network Functions Virtualisation into the telecommunications infrastructure can become the basis for improving the quality of services, optimising costs, and ensuring a high level of security in the context of digital transformation and infrastructure modernisation, are derived |
|---|---|
| ISSN: | 19999941 20786387 |
| DOI: | 10.63341/vitce/1.2025.124 |