Join us on Zoom: https://ucsc.zoom.us/j/94826219208?pwd=NGtzRWx1dy92SWxYN0tVeVpGWEVIZz09
Abstract: Data processing is growing enormously. From deep neural networks with trillions of parameters to graph processing with trillions of edges and vertices, the amount of memory required for modern processing can easily be at the terabyte scale. Unfortunately, traditional data movement and data security techniques are inefficient or inappropriate at this scale. In this talk, I'll describe a set of hardware-software co-design techniques to overcome the inefficiencies of both moving and securing data.
Bio: Jason Lowe-Power is an assistant professor of Computer Science at the University of California, Davis. His research targets important end-to-end applications (e.g., big-data analytics) and develops new hardware, software, and systems to improve their performance increase their scalability. He received his Ph.D. from University of Wisconsin-Madison Computer Sciences department in Summer 2017 and his M.S. in Summer 2013. He also received a B.S. in Computer Science from Georgia Tech in 2010.