Real-time networks and Software-Defined Networking (SDN) play pivotal roles in contemporary communication systems, supporting applications such as video conferencing, online gaming, VoIP calls, virtual reality (VR) systems, and Industry 4.0 automation. These applications necessitate low and predictable latency for real-time interactivity, responsiveness, and uninterrupted performance. Achieving low and predictable latency in dynamic network environments is challenging, as factors like changing network topology, traffic load, and connected devices impact latency. Furthermore, network protocols and devices introduce additional delays. SDN provides a solution by allowing dynamic control of communication paths, enhancing the performance of time-sensitive applications. This paper addresses the crucial task of evaluating latency in real-time networks. We propose a novel measurement methodology that effortlessly evaluates latency by utilizing Time-Sensitive Networking (TSN) features and tools embedded in the Linux kernel. Our methodol- ogy accurately captures network latency irrespective of workload, configuration, and computational capabilities.
A TSN-based Technique for Real-Time Latency Evaluation in Communication Networks
Morato, Alberto
Primo
;Zunino, ClaudioSecondo
;Cheminod, Manuel;Vitturi, StefanoPenultimo
;Tramarin, FedericoUltimo
2024
Abstract
Real-time networks and Software-Defined Networking (SDN) play pivotal roles in contemporary communication systems, supporting applications such as video conferencing, online gaming, VoIP calls, virtual reality (VR) systems, and Industry 4.0 automation. These applications necessitate low and predictable latency for real-time interactivity, responsiveness, and uninterrupted performance. Achieving low and predictable latency in dynamic network environments is challenging, as factors like changing network topology, traffic load, and connected devices impact latency. Furthermore, network protocols and devices introduce additional delays. SDN provides a solution by allowing dynamic control of communication paths, enhancing the performance of time-sensitive applications. This paper addresses the crucial task of evaluating latency in real-time networks. We propose a novel measurement methodology that effortlessly evaluates latency by utilizing Time-Sensitive Networking (TSN) features and tools embedded in the Linux kernel. Our methodol- ogy accurately captures network latency irrespective of workload, configuration, and computational capabilities.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.