This book introduces GPU computing using CUDA from the contemporary perspective of the Julia programming language. It is designed to empower scientists, engineers, and computational researchers to harness the massive parallelism of NVIDIA GPUs for solving high-performance numerical problems.

Readers will gain both theoretical and practical understanding of GPU architectures, CUDA kernels, memory hierarchies, and performance tuning strategies using the expressive and high-level syntax of Julia.

Main Emphasis:

• Practical GPU programming in Julia using CUDA.jl
• Deep integration of Julia’s scientific computing ecosystem with GPU workloads
• Performance optimization, memory tuning, and debugging techniques
• Real-world examples from numerical computing, image and signal processing, simulation, and machine learning

Arnis Lektauers is an Associate Professor and Senior Researcher at the Institute of Information Technology of Riga Technical University, Latvia. His research focuses on high-performance interactive hybrid simulation and its application to the analysis of complex systems in industrial, economic, ecological and sustainable development contexts.