In a world that is rapidly moving to many core, not multi core, CUDA programming model presents a different, in some ways refreshing, approach to expressing parallelism. CUDA is the means by which software engineers use to communicate and instruct GPUs to handle data and work in tandem with the CPU. GPU devotes much more transistors to data processing rather than data caching and flow control.
This whitepaper shows how applications can be accelerated using CUDA to achieve significant parallel speedups. Such applications fall into a variety of problem domains, including machine learning, database processing, bioinformatics, financial modeling, numerical linear algebra, medical imaging, and physical simulation, among others.
Download attachmentSelect Internal Page:
