In the new Software-Defined Networking (SDN) paradigm some challenges such as suitable traffic load distribution strategies, which can negatively affect the network performance, are faced. In this paper, a Multi-Resource Aware Load-Balancing strategy exploiting Software Defined Networking (SDN) programmability is presented. Our approach, leveraging the SDN controller, distributes the web requests to a web server cluster according to four metrics: CPU, Memory, Network Traffic, and Response Time of each web server. Through a comprehensive experimental campaign, we evaluated and assessed the advantages of the proposed load-balancing approach against basic methods like random selection and round-robin and also against recent advancements in load-balancing strategies that are considered more sophisticated and intelligent.
MuRA-LB: A Multi-Resources Aware Load-Balancing Strategy Exploiting Software-Defined Networking
Spina M. G.;Tropea M.;De Rango F.
2025-01-01
Abstract
In the new Software-Defined Networking (SDN) paradigm some challenges such as suitable traffic load distribution strategies, which can negatively affect the network performance, are faced. In this paper, a Multi-Resource Aware Load-Balancing strategy exploiting Software Defined Networking (SDN) programmability is presented. Our approach, leveraging the SDN controller, distributes the web requests to a web server cluster according to four metrics: CPU, Memory, Network Traffic, and Response Time of each web server. Through a comprehensive experimental campaign, we evaluated and assessed the advantages of the proposed load-balancing approach against basic methods like random selection and round-robin and also against recent advancements in load-balancing strategies that are considered more sophisticated and intelligent.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


