We show you the long-awaited way to learn CUDA without NVIDIA GPUs!
Finally, you can learn CUDA on your laptop, tablet, or even your cell phone, and that’s all there is to it! You can use CUDA to write general-purpose programs that take advantage of the huge computational power of modern GPUs, as well as powerful libraries for machine learning, image processing, linear algebra, and other things.
What do you learn from this?
With Docker and OS-level virtualization, we’ll show you how to learn CUDA with simple tools like GPGPU-Sim, which is a cycle-level simulator of modern graphics processing units (GPUs) running GPU computing tasks written in CUDA or OpenCL. This course is meant to help you learn about NVIDIA’s CUDA parallel architecture and programming model in a way that is easy to understand. We want to keep the lessons up to date and add new lessons and exercises every month!
- Virtualization: The basics.
- Essentials of Docker.
- This is a guide to the basics of the GPU
- CUDA is installed.
- The CUDA Toolkit
- CUDA threads and blocks can be used in different ways.
- Examples of CUDA coding
Zoom live class lectures are now part of the course, and we’re going to show you how to work with parallel and distributed computing and High-Performance Computing (HPC) systems. Slurm, PBS Pro, OpenMP, and CUDA are all part of the software stack. People will be able to take live classes through the Scientific Programming School, which is an interactive and advanced e-learning platform for learning how to write scientific code. Students who buy this course will get free access to the interactive version of this course from the Scientific Programming School. This version has Scientific code playgrounds and is more interactive (SCIENTIFIC PROGRAMMING IO). If you look in the bonus content, you’ll see how to join.
People who own NVIDIA own some of the images used in this class, so they can’t use them.