About the Book

CUDA

CUDA (Compute Unified Device Architecture) is a hardware architecture and programming model introduced by NVIDIA for the parallel processing of Graphics Processing Units (GPUs).

WHAT'S POSSIBLE WITH CUDA

Thanks to CUDA, developers and researchers can write industrial and scientific applications exploiting the massive parallelism released by the thousands of cores of NVIDIA graphic cards using extensions of common programming languages like C/C++, Java, Fortran, Python, and Matlab or accelerated libraries as OpenCV, Tensorflow, and Pytorch. CUDA-based GPU acceleration has permitted, after 10 years from its very first launch, the development of applications considered severely computationally demanding or even impossible to be run. These have regarded biomedical imaging, oil and natural resource exploration, computer vision, bioinformatics, and life science until opening the doors to artificial intelligence today. We are all astonished when a web application is capable to recognize the face of our friends in a photo, but few of us are aware that CUDA is running in the background.

WHAT WE WILL DO IN THIS BOOK

In this book, we will see CUDA, in its various forms, at work on modern applications of science and technology which require high-performance processing.

We will catch the CUDA typical usage along with the related programming tricks and exploit the largely available CUDA environment made of several, highly optimized and performing libraries.

THE CODES

We will develop several codes related to modern applications of parallel programming with CUDA. All the developed codes will be available for download. In this book, different projects will be presented which will regard the application of parallel programming on GPU to problems of industrial and scientific interest, by using CUDA as the programming language.

WORK IN PROGRESS

Book writing is still in progress.

About the Author

Evgenij Lebedev

I’m teaching, researching and consulting parallel programming on Graphics Processing Units (GPUs) since the delivery of CUDA. I also play Matlab and Python.