Large computational problems are offloaded onto a GPU because the problems run substantially faster on the GPU than on the CPU. By leveraging the innate parallelism of the GPU overall performance of the application is improved. (For example, see here and here.) However a second collateral benefit of moving computation to the GPU is the…
CenterSpace Software is pleased to announce the general availability of the Premium Editions of NMath 5.3 and NMath Stats 3.6. NMath Premium leverages the power of NVIDIA’s CUDA™ architecture for GPU-accelerated mathematics in .NET. CenterSpace has partnered with leading GPU computing experts to bring GPU-accelerated mathematics to the .NET platform for the first time.
NMath Premium was designed to provide an easy to follow pathway for .NET developers to leverage the performance of the GPU without having to wade through the complexities of GPU programming and their attendant details. And to provide a GPU-aware math library that can be used by developers to build once & run anywhere without concerning themselves of their user’s installed GPU model, version, or even the existence of an GPU.
NMath Premium is designed to be a near drop-in replacement for NMath. However, there are a few important configuration differences and additional logging capabilities that are specific to the premium product.
NMath Premium is CenterSpace Software’s NVIDIA GPU-accelerated edition of the NMath math and statistics library. Many linear algebra and signal processing algorithms can now run on a local NVIDIA GPU processor, frequently realizing several multiples of performance gain. In this post, we look at the performance of complex to complex forward 1D and 2D FFT’s.