I am using CUDA for a very large data computation , mainly for loops are used. I need maximum efficiency and least possible time!
Since i am a beginner in CUDA computation i don’t know how to compute CUDA in c# project- i tried googling but not getting any proper instructions that i might understand as a beginner!
Hi Sifatul,
I am not an expert, I started with cuda 1 week ago, but I recommend that you read the ‘Cuda c programming guide’ (http://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html), it is a good start point for begginers. Also you can read the following article https://devblogs.nvidia.com/parallelforall/easy-introduction-cuda-c-and-c/. This article is part of a series of articles with CUDA.
I think that there are not a proper instructions for started with CUDA, the best option is read an read and experiment yourself. The good notices is that there is many information in the official pages.
Some people who want to use CUDA in C# use Cudafy or managedCuda.
I usually make wrapper with C++/CLI for bridge between C#(.NET) and C/C++
I use managedCUDA to run a CUDA C/C++ code through C# application. If you’re beginning in CUDA I recommend that you read the official documentation and this book : https://www.physics.drexel.edu/~valliere/PHYS405/GPU_Story/CUDA_by_Example_Addison_Wesley_Jul_2010.pdf