This talk will cover NVIDIA's latest and fastest GPU called Maxwell, pointing out its architectural advantages. The talk will also give a comparison of serial and parallel algorithms, and the assumptions therein. It will include discussions on how theoretical parallel algorithms map to parallel hardware, some limitations and solutions, and the unquestioning reliance on asymptotic analysis for choosing algorithms.

The talk represents a collection of work that Dr. Lars Nyland have done while at NVIDIA for the past 10 years, as an architect in the GPU division.

Dr. Lars Nyland is a senior architect in the GPU division at NVIDIA. Prior to joining NVIDIA, he was a professor of computer science at the Colorado School of Mines and UNC Chapel Hill.