Please note: This seminar will take place in DC 1304.
Niloy Saha, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Raouf Boutaba
Efficient management of network slices in 5G and beyond networks relies heavily on data-driven algorithms, which require timely and accurate network monitoring. However, monitoring network slices end-to-end (E2E) presents significant challenges. Current solutions are often fragmented, lacking the ability to correlate network-wide data across multiple segments. Furthermore, the diverse requirements of network slices for Key Performance Indicators (KPIs) and monitoring granularity demand dynamic, real-time adjustments to minimize resource overhead without compromising accuracy.
In this talk, I will present Monarch, a scalable, cloud-native monitoring architecture tailored for 5G deployments. Monarch addresses the challenges of E2E network slice monitoring by enabling efficient KPI computation per slice. Our experiments on a 5G network slice testbed using real-world datasets show that Monarch achieves consistent data ingestion times of 2.25–2.75 ms and reduces monitoring overhead by up to 76% while maintaining accuracy.