GPGPU (General Purpose computation on Graphics Processing Unit) has emerged as the necessary skill for anyone who writes computer software. With hundreds of millions of GPUs already in the market, programming them is a valuable skill that can open new opportunities and avenues. We conduct fast-track workshops on how to program GPUs for amateur as well as advanced students and faculty. These workshops are conducted by our NVIDIA certified GPU experts who developed CUVI – an award-winning GPU accelerated Computer Vision library and regularly work on complex problems to accelerate them on GPUs. Here’s what we currently offer:

1. Two day workshop on GPGPU Programming for beginners

This 2 day workshop covers the basics of GPGPU, covering the existing software frameworks available and a hands-on development using NVIDIA CUDA and OpenCL. By the end of this workshop, attendees will get a grasp of what GPGPU is all about, basic concepts of how to write parallel code and be able to write simple programs on GPUs.

2. Five-day Short course on GPGPU using NVIDIA CUDA

This short course will cover the essential topics in detail for the first 3 days after which the attendees will be taught on how to translate complex problems from the serial into parallel domain and how to make the best use of GPU resources. By the end of this course, attendees will have a fair idea of the architecture of a modern GPU and how to exploit it to write optimized code for different generations of NVIDIA CUDA enabled GPUs available.