%0 Journal Article
%T GPGPU PROCESSING IN CUDA ARCHITECTURE
%A Jayshree Ghorpade
%A Jitendra Parande
%A Madhura Kulkarni
%A Amit Bawaskar
%J Advanced Computing : an International Journal
%D 2012
%I Academy & Industry Research Collaboration Center (AIRCC)
%X The future of computation is the Graphical Processing Unit, i.e. the GPU. The promise that the graphicscards have shown in the field of image processing and accelerated rendering of 3D scenes, and thecomputational capability that these GPUs possess, they are developing into great parallel computingunits. It is quite simple to program a graphics processor to perform general parallel tasks. But afterunderstanding the various architectural aspects of the graphics processor, it can be used to performother taxing tasks as well. In this paper, we will show how CUDA can fully utilize the tremendous powerof these GPUs. CUDA is NVIDIA¡¯s parallel computing architecture. It enables dramatic increases incomputing performance, by harnessing the power of the GPU. This paper talks about CUDA and itsarchitecture. It takes us through a comparison of CUDA C/C++ with other parallel programminglanguages like OpenCL and DirectCompute. The paper also lists out the common myths about CUDAand how the future seems to be promising for CUDA.
%K GPU
%K GPGPU
%K thread
%K block
%K grid
%K GFLOPS
%K CUDA
%K OpenCL
%K DirectCompute
%K data parallelism
%K ALU
%U http://airccse.org/journal/acij/papers/0112acij09.pdf