GPU Computing SDK 4.1.21 RC 2

The NVIDIA GPU Computing SDK provides hundreds of code samples, white papers, to help you get started on the path of writing software with CUDA C/C++, OpenCL or DirectCompute.<br /><br />The SDK code samples covers a wide range of applications and techniques, including:<br /> <br /><br /> Simple techniques demonstrating image<br /> Basic approaches to GPU Computing<br /> Best practices for the most important features<br /> Working efficiently with custom data types<br /> Quickly integrating GPU acceleration into C and C++ applications<br /> How-To examples covering topics such as:<br /> Adding support for GPU-accelerated libraries to an application<br /> Using features such as Zero-Copy Memory, Asynchronous Data Transfers, Unified Virtual Addressing, Peer-to-Peer Communication, Concurrent Kernels, and more<br /> Sharing data between CUDA and Direct3D/OpenGL graphics APIs (interoperability)<br /> image Data-parallel algorithms and primitives for linear algebra operations:<br /> Matrix transpose<br /> Matrix-matrix multiplication<br /> Matrix multiplication with multiple right hand sides<br /> Parallel prefix sum of large arrays<br /> Any many more!<br /> Performance measurement and optimization<br /> Bandwidth tests<br /> Application profiling using timers<br /> imageAdvanced application examples<br /> Using CUDA with MPI and OpenMP<br /> Computational fluid dynamics (CFD)<br /> Gravitational n-body simulation<br /> Black-Scholes & binomial option pricing<br /> 3D Finite-difference time-domain (FDTD)<br /> Video encode/decode<br /> Image convolution<br />

Requirements:	No special requirements
Platforms:	Mac, Mac OSX
Keyword:	Arrays,  Large,  Matrix,  Matrixmatrix,  Multiple,  Multiplication,  Performance,  Sides,  Transpose
Users rating:	0/10