[Abstract]

Prediction qAueries are widely used across industries to perform advanced analytics and draw insights from data. They include a data processing part (e.g., for joining, filtering, cleaning, featurizing the datasets) and a machine learning (ML) part invoking one or more trained ML models to perform predictions. These two parts have so far been optimized in isolation, leaving significant opportunities for optimization unexplored. In this talk, I will present Raven, a production-ready system for optimizing prediction queries holistically across data and ML part. I will also briefly explore how hardware accelerators (e.g., GPUs and FPGAs) can improve the performance of such queries.

[Biography]

Kwanghyun Park is an Assistant Professor at Yonsei University, looking at system support for machine learning workloads and system optimizations with machine learning techniques. He is also interested in query processing and optimizations with new classes of hardware. Prior to his current role, he was a senior research engineer at Microsoft Gray Systems Lab (GSL). He obtained his Ph.D. and M.Sc. in Computer Science from University of Wisconsin-Madison (database research group) under the supervision of Prof. Jignesh M. Patel after receiving a Bachelor with summa cum laude in Computer Science and Applied Math & Statistics (AMS) from State University of New York, Stony Brook.