搜索资源列表
GPUFIR
- 一个基于GPU运算的FIR滤波器程序,基于CUDA平台而开发。支持GPU并行运算。-GPU-based computing a FIR filter process, developed based on the CUDA platform. Support GPU parallel computing.
asyncAPI
- CUDA开发计算,异步计算接口。可以在主机与现存之间异步执行。来自NVIDIA SDK-CUDA development of computing, asynchronous computing interface. Can be asynchronous between the host and the existing implementation. From the NVIDIA SDK
NVIDIA_CUBLAS_1.1
- Nividia提供的CUDA的BLAS库源码,可以在GPU上运行BLAS函数-Nividia provided CUDA source code of the BLAS library, you can run GPU function BLAS
cksc2.0
- nvidia CUDA 2.0 SDK 中文版本,翻译率大概为95%-nvidia CUDA 2.0 SDK Chinese version, the translation rate of about 95
matrixMul
- 采用GPU通用计算API(CUDA)实现矩阵相乘。-The use of general-purpose GPU computing API (CUDA) to achieve the matrix multiplication.
MonteCarlo
- 采用GPU通用计算API(CUDA)实现蒙特卡罗方程。-The use of general-purpose GPU computing API (CUDA) the realization of the Monte Carlo equation.
ComputeTime
- 基于NV的CUDA(计算统一构架)的一段程序,示例如何在显卡的GPU上进行并行平方和运算,代码思路清晰,易于理解。-NV based on the CUDA (calculated unified fr a mework) period of procedures, examples of how the GPU graphics cards on the square and parallel computing, code clea
fft2d_gpu
- using NVIDIA FFT CUDA Library to solve FFT problem
RT1
- 使用CUDA进行并行光线跟踪的原代码,需要支持CUDA的显卡和相应驱动-The use of CUDA parallel ray tracing of the original code, the need to support CUDA graphics card and the corresponding drive
ANeuralNetworkonGPU
- CUDA应用:在CUDA上实现神经网络,识别手写数字-the implementation of a neural network with CUDA. Neural Network for Recognition of Handwritten Digits
BFS_CUDA
- 用CUDA实现BFS算法源码。CUDA是NV公司基于GPU的统一计算架构,BFS复杂度由CPU上的O(V+E)降为O(diameter),直径diameter即图中root到leaf的最长距离。-This is the algorithm of BFS(breadth first search) on CUDA, and its time complexity is down to O(diameter) which diameter
UIUC_CudaProgrammingGuide
- UIUC的工程院在全美堪称至尊级,这是UIUC推出的CUDA学习教程,非常系统地讲述其特点及应用。很好,很强大!-UIUC' s at the National Academy of Engineering called Extreme class, This is UIUC Study launched CUDA tutorial, very systematically about their characteristics
Mars_v2
- GPU实现的MapReduce fr a mework,对于学习并行编程和cuda平台的编程方面有着极好的参考价值,里面附带论文。用户要求有NViDIA显卡,并且安装cuda编程环境。-We design and implement Mars, a MapReduce fr a mework, on graphics processors (GPUs). MapReduce is a distributed programming
dct8x8
- cuda开发环境下的矩阵运算,cuda sdk中的内容,矩阵运算充分显示了并行编程的优势,对于学习cuda编程有着很好的借鉴作用-matrix operations in parallel with cuda platform
transpose
- cuda平台开发的矩阵变换并行操作(Matrix transpose),属于cuda sdk中的内容-Small Matrix transpose with Cuda (Example for a 16x16 matrix)
cuda_hist
- cuda compute unified device architecture histogram code
mv
- cuda Faster Matrix-Vector Multiplication on GeForce 8800GTX
SumProductGPU.tar
- cuda product of sums algorithm on the gpu
CUDA_lecture
- 这是Nvidia CUDA基础知识演讲的幻灯片-This is a powerpoint about Nvidia CUDA
cd40192
- 清华大学cuda培训教程 gpu并行算法 高清版pdf-Tsinghua University training course cuda parallel algorithm gpu HD version pdf