Please note: This PhD seminar will take place in DC 1304.
Niloy Saha, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Raouf Boutaba
Detecting QoS anomalies in 5G user planes requires fine-grained per-flow visibility, but existing telemetry approaches face a fundamental trade-off. Coarse per-class counters are lightweight but mask transient and flow-level anomalies, while per-packet telemetry postcards provide full visibility at a prohibitive cost that grows linearly with line rate. Selective postcard schemes reduce overhead but miss anomalies that fall below configured thresholds or occur during brief intervals.
In this talk, we present Kestrel, a sketch-based telemetry system for 5G user planes that provides fine-grained visibility into key metric distributions such as latency tails and inter-arrival times at a fraction of the cost of per-packet postcards. Kestrel extends Count-Min Sketch with histogram-augmented buckets and per-queue partitioning, which compresses per-packet measurements into compact summaries while preserving anomaly-relevant signals. We show that incorporating sketch collisions into detectability guarantees gives principled guidance for sketch sizing and binning, preserving anomaly separability. Our evaluations on a 5G testbed with Intel Tofino switches show that Kestrel achieves 10% better detection accuracy than existing selective postcard schemes while reducing export bandwidth by 10×.