A map reduce framework for programming graphics processors
Bryan Catanzaro, Narayanan Sundaram, Kurt Keutzer · 2010
Recent developments in programmable, highly parallel Graphics Processing Units (GPUs) have enabled high performance general purpose computation. We describe a framework designed for high performance GPU programming, built on Nvidia’s Compute Unified Device Architecture (CUDA) platform. The framework is built around the Map Reduce abstraction, which allows application developers to focus on their application, while enabling high performance GPU implementation. We show the utility of our framework by implementing Support Vector Machine training as well as classification, achieving speedups of up to 32 × and 150 × respectively over commonly used SVM software running on a CPU. 1.