Alan (Zaoxing) Liu
Title: Universal Monitoring
Host: Minlan Yu
In network measurement, obtaining accurate estimates of multiple metrics given router CPU and memory constraints is a challenging problem. Existing approaches fall in one of two undesirable extremes: low fidelity general purpose approaches such as sampling, or high fidelity but complex algorithms customized to specific application level metrics. In this talk, I will present framework for flow monitoring which leverages the techniques of Universal Sketches in the streaming model and demonstrates that it is possible to achieve both generality and high accuracy on estimating metrics. UnivMon uses an application-agnostic data plane monitoring primitive; different (and possibly unforeseen) estimation algorithms run in the control plane, and use the statistics from the data plane to compute application-level metrics. We evaluate the effectiveness of UnivMon using a range of trace-driven evaluations and show that it offers comparable (and better) accuracy relative to custom sketching solutions.
Zaoxing Liu is a PhD student in the Department of Computer Science at Johns Hopkins University. His research interests include SDN, network measurement, data streaming algorithms, and “big data” processing systems using streaming algorithms. His previous work includes theoretical results in the data streaming model and effective systems in network monitoring and cosmological simulation. He developed several open-source software packages for streaming data analysis.