GPU Accelerated Computing with C and C++

Using the CUDA Toolkit you can accelerate your C or C++ applications by updating the computationally intensive portions of your code to run on GPUs. To accelerate your applications, you can call functions from drop-in libraries as well as develop custom applications using languages including C, C++, Fortran and Python. Below you will find some resources to help you get started using CUDA.

1
SETUP CUDA

Install the free CUDA Tookit on a Linux, Mac or Windows system with one or more CUDA-capable GPUs. Follow the instructions in the CUDA Quick Start Guide to get up and running quickly.

Or, watch the short video below and follow along.

If you do not have a GPU, you can access one of the thousands of GPUs available from cloud service providers including Amazon AWS, Microsoft Azure and IBM SoftLayer. The NVIDIA-maintained CUDA Amazon Machine Image (AMI) on AWS, for example, comes pre-installed with CUDA and is available for use today.

For more detailed installation instructions, refer to the CUDA installation guides. For help with troubleshooting, browse and participate in the CUDA Setup and Installation forum.

2
YOUR FIRST CUDA PROGRAM

You are now ready to write your first CUDA program. The article, Even Easier Introduction to CUDA, introduces key concepts through simple examples that you can follow along.

The video below walks through an example of how to write an example that adds two vectors.

The Programming Guide in the CUDA Documentation introduces key concepts covered in the video including CUDA programming model, important APIs and performance guidelines.

NVIDIA also provides hands-on training through a collection of self-paced labs. The labs guide you step-by-step through editing and execution of code, and even interaction with visual tools is all woven together into a simple immersive experience.

3
PRACTICE CUDA

Practice the techniques you learned in the materials above through more hands-on labs created for intermediate and advanced users.

The CUDA C Best Practices Guide presents established parallelization and optimization techniques and explains programming approaches that can greatly simplify programming GPU-accelerated applications.

For a more formal, instructor-led introduction to CUDA, explore the Introduction to Parallel Programming on UDACITY. The course covers a series of image processing algorithms such as you might find in Photoshop or Instagram. You'll be able to program and run your assignments on high-end GPUs, even if you don't have one yourself.

Additional Resources

CODE Samples

Availability

The CUDA Toolkit is a free download from NVIDIA and is supported on Windows, Mac, and most standard Linux distributions.

So, now you’re ready to deploy your application?
Register today for free access to NVIDIA TESLA GPUs in the cloud.

Latest News

7 Things You Might Not Know about Numba

Numba is a Python compiler from Anaconda that can compile Python code for execution on CUDA-capable GPUs or multicore CPUs.

Style Transfer From Multiple Reference Images

Researchers from Microsoft and Hong Kong University of Science and Technology developed a deep learning method that can transfer the style and color from multiple reference images onto another photograph.

Webinar: Build Your Next Deep Learning Application for NVIDIA Jetson in MATLAB

Learn how you can use MATLAB to build your computer vision and deep learning applications and deploy them on NVIDIA Jetson.

PGI 17.7 Delivers OpenACC and CUDA Fortran for Volta GPUs

PGI compilers & tools are used by scientists and engineers who develop applications for high-performance computing (HPC) systems.

Blogs: Parallel ForAll

Programming Tensor Cores in CUDA 9

A defining feature of the new Volta GPU Architecture is its Tensor Cores, which give the Tesla V100 accelerator a peak throughput 12 times the 32-bit floating point throughput of the previous-generation Tesla P100.

Register Cache: Caching for Warp-Centric CUDA Programs

In this post we introduce the “register cache”, an optimization technique that develops a virtual caching layer for threads in a single warp. It is a software abstraction implemented on top of the NVIDIA GPU shuffle primitive.

Mixed-Precision Training of Deep Neural Networks

Deep Neural Networks (DNNs) have lead to breakthroughs in a number of areas, including image processing and understanding, language modeling, language translation, speech processing, game playing, and many others.

Training AI for Self-Driving Vehicles: the Challenge of Scale

Modern deep neural networks, such as those used in self-driving vehicles, require a mind boggling amount of computational power.