tag:blogger.com,1999:blog-71918460718529785762024-10-24T18:46:19.813-07:00Alex Slesarenko's blogAlex Slesarenkohttp://www.blogger.com/profile/10253205349391319347noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-7191846071852978576.post-12744500026598601282009-06-23T13:06:00.000-07:002009-06-28T12:16:34.848-07:00CUDA programming in F#: Part 2, Accelerating Mandelbrot sampleIn the previous post (CUDA programming in F#: Part 1), we talked about how to setup dev environment to get started with CUDA programming in F#. BitonicSort sample used there is a classical parallel sorting algorithm, with non trivial GPU kernel (taken from CUDA SDK) . Unfortunately, it's not good enough to evaluate benefits of GPU acceleration as it has execution time comparable to CPU.This timeAlex Slesarenkohttp://www.blogger.com/profile/10253205349391319347noreply@blogger.com1tag:blogger.com,1999:blog-7191846071852978576.post-66103197193577414302009-05-15T14:27:00.000-07:002009-06-28T02:59:01.994-07:00CUDA programming in F#: Part 1, Getting startedIf you develop computationaly intensive applications, you will probably benefit from massive parallelization on CUDA devices. CUDA is a technology that can be used to harness full power of NVIDIA GPU device(s) installed on usual computer (server or desktop) for general purpose computations. Many applications can be parallelized using data-parallel algorithms and executed on CUDA enabled devices Alex Slesarenkohttp://www.blogger.com/profile/10253205349391319347noreply@blogger.com3