Traffic-Specific Packet-Level Monitoring in SDN

Software-Defined Networking (SDN) has transformed network administration, highlighting the importance of monitoring techniques in improving network performance and reliability. There is a need for flexible monitoring tools that provide adaptive, real-time network behavior insights while trying to minimize error detection time. This paper addresses the task of exploring the adaptability of a monitoring tool called AdaptMon (Adaptive Monitoring), which results in a new monitoring algorithm called Traffic-Specific Packet-Level (TSPL) AdaptMon. Depending on the requirements of the network, AdaptMon may switch between general, switch-level monitoring or detailed, packet-level monitoring. The object of this paper is to investigate the application of AdaptMon's adaptation capabilities in the domain of video streaming, where the Quality of Service (Qo s) is directly impacted by network performance. Our findings show that customizing AdaptMon to address problems associated with Real-time Transport Protocol (RTP) traffic, where the monitored variable is jitter, can positively affect monitoring processing time and error detection delay. TSPL AdaptMon achieves a 66.6 % improvement in processing delay and a 16.6 % improvement in the cumulative performance metric compared to the General Switch-Level AdaptMon method. The implications of these findings are that variants of the AdaptMon monitoring protocol can dynamically modify the monitoring rate to improve the insights obtained about the SDN's condition.