Conference paper Open Access
Marios Touloupou; Evgenia Kapassa; Athanasios Kiourtis; Panayiotis Stavrianos; Dimosthenis Kyriazis
Following the rapid development and innovations of mobile networking technologies, an entirely new era of mobile communications systems is evolving, namely the fifth-generation mobile technology (5G). Among the goals of 5G networks, one is to meet predetermined Quality of Service (QoS) requirements, in different application scenarios (e.g., such as data transmission rate and latency). To this end, the demand for always-accessible applications and services is continuously growing, rendering the monitoring of network availability a fundamental requirement for modern infrastructures. Network monitoring systems provide a first level of reliability in case of application failure or performance deterioration. Meanwhile, a monitoring framework should accumulate the process and deliver the monitored data at every requested aggregation level and frequency, without introducing a substantial overhead into the system. In this paper, we focus on the trade-off between the running services and the monitoring framework itself. We propose an adaptive scheduling algorithm, aiming at the removal of static time interval applied in the monitoring system. The main goal is to provide highly accurate information in real time without producing unnecessary traffic to the network.