Newsletter

Newsletter
 

GPU Whitepaper Library

 


        

ArrayFire Whitepaper:
High-level GPU Computing in C, C++, and Fortran

Login to Download

ArrayFire is a software platform developed at AccelerEyes that allows users and programmers to rapidly develop data-parallel programs in C, C++, and Fortran. ArrayFire provides a simple, high-level matrix abstraction over low-level GPU APIs such as CUDA, OpenCL, and OpenGL along with thousands of GPU-tuned functions to allow users in science, engineering, and finance to take full advantage of GPU hardware. The combination of an easy-to-use array interface, automatic memory management, on-the-fly compilation, parallel GPU FOR-loop construction, and interactive hardware-accelerated graphics library, make ArrayFire well suited to rapid prototyping of data-parallel algorithms and building end-to-end applications.





        

Matrix Computations on the GPU with ArrayFire
for C/C++, by Andrzej Chrzȩszczyk

Login to Download

In recent years the Graphics Processing Units (GPUs) designed to efficiently manipulate computer graphics are more and more often used to General Purpose computing on GPU (GPGPU). NVIDIA's CUDA and OpenCL platforms allow for general purpose parallel programming on modern graphics devices. Unfortunately many owners of powerful graphic cards are not experienced programmers and can find these platforms quite difficult. The purpose of this document is to make the first steps in using modern graphics cards to general purpose computations simpler. In the first two chapters we want to present the ArrayFire software library which in our opinion allows to start computations on GPU in the easiest way. In the present text we describe the ArrayFire free version. It allows for efficient dense matrix computations in single precision on a single GPU. The readers interested in double precision linear algebra, multiple GPUs, or sparse matrices should consider ArrayFire Pro.