Please note: This seminar will take place in DC 1304.
Qizhen Zhang, Assistant Professor
Department of Computer Science, University of Toronto
Cloud data centers are undergoing once-in-a-decade transformation as they evolve to house hyperscale services and AI workloads. What opportunities does this create for database systems?
In this talk, I will explore this question through our recent work on data processing with DPUs and GPUs. I will first introduce DPDPU, a platform for building low-cost high-performance database systems using DPUs, and the latest progress on optimizing storage (with DDS) and compute (with dpKernels) under this umbrella. I will also present MGI, a communication interface for databases to access massive GPU infrastructures to unlock unprecedented scale of database acceleration.
Bio: Qizhen Zhang is an Assistant Professor of Computer Science at the University of Toronto, where he leads the Far Data Lab. His current research focuses on building infrastructures for large-scale data processing, and he is broadly interested in data management and computer systems and networking. His research appears at top-tier database and systems/networking conferences such as SIGMOD/VLDB and SIGCOMM/NSDI.
He received his Ph.D. in the Department of Computer and Information Science at the University of Pennsylvania, where his work was recognized with the best Computer Science Ph.D. dissertation award. He spent a gap year in industry at Microsoft Research, Redmond, before joining UofT. His lab has been supported by Connaught New Researcher Award, Digital Research Alliance of Canada Resources for Research Groups Grant, NSERC Discovery and Alliance grants, and gifts from industrial sponsors such as Amazon and Google.
More details can be found on Qizhen Zhang’s research site and the Far Data Lab site.