NVIDIA Developer Zone

Get Started - Parallel Computing

Get started quickly with GPU Computing using the solution that best meets your needs. Your options include simply dropping in a GPU-accelerated library, adding a few GPU Directives in your code, or designing your own parallel algorithms.  And you can combine these approaches to accelerate your applications:

  GPU-Accelerated Libraries
Drop in a GPU-accelerated library to replace MKL, IPP, FFTW and other widely-used libraries		  
 
C++ Template Library
cuBLAS                              cuSPARSE
Linear Algebra
NPP                                      cuFFT
Signal & Image Processing		  
More GPU-Accelerated Libaries

 
  GPU Directives
Automatically parallelize loops in your Fortran or C code using OpenACC directives		  
 
  • Easy : simply insert hints in your code
  • Open : run on either CPU or GPU
  • Powerful: tap into the power of GPUs within minutes
Learn More...
  
Learn More About Directives
  Programming Languages
Develop your own parallel applications and libraries using a programming language you already know		  
 

 CUDA C/C++
GPU Acceleration for 
C and C++ Apps.
Learn more...

 CUDA Fortran
GPU Acceleration for
Fortran Applications
Learn more...

  
More Programming Language Solutions

Looking for more? Learn more about GPU-accelerated applications, tools and libraries