Dynamic data structures allow updates to the datastructure without having to rebuild it completely. I'll discuss five dynamicGPU data structures that we designed and built—log-structuredmerge trees, quotient filters, linked lists and hash tables built atop them,dynamic graphs, and B-trees. The talk is less about the specifics of these datastructures and more about principles that we followed in building them and whatwe learned from the experience.
John Owens is the Child Family Professor of Engineering and Entrepreneurship in the Department of Electrical and Computer Engineering at the University of California, Davis, where he leads an amazing group of graduate students in a research program centered around parallel computing on the graphics processing unit (GPU). He earned his PhD from Stanford in 2003 and his BS from UC Berkeley in 1995. His first and only visit to Beijing was in 2007.