Modern computing has been transformed by the GPU — and mastering it has become one of the most valuable and sought-after skills in software engineering, scientific computing, artificial intelligence, and systems design. This series gives you the complete, uncompromising technical education needed to understand, program, and fully exploit the power of modern graphics processing units at every level of the stack.
Beginning with a thorough grounding in GPU hardware architecture — how modern GPUs are designed, how execution units, memory hierarchies, schedulers, and interconnects actually function under the hood — the series builds systematically toward the highest levels of parallel programming expertise. You won't just learn to write GPU code; you'll learn to write GPU code that performs. Understanding the architecture beneath your programs is what separates engineers who get acceptable results from engineers who push hardware to its absolute limits.
From low-level assembly programming that exposes the raw instruction set and microarchitectural behaviour of the GPU, to high-level CUDA development with advanced optimization strategies, the series covers the full breadth of GPU programming with the depth that professional engineers and researchers demand. Topics span thread hierarchy and memory models, warp-level execution, occupancy tuning, kernel fusion, asynchronous execution, memory coalescing, and the architectural features that define the latest GPU generations.
Parallel computing volumes tie everything together — demonstrating how architectural knowledge and programming expertise combine to solve real, large-scale computational problems across domains including simulation, machine learning infrastructure, image processing, scientific computing, and high-performance data pipelines.
Whether you are a systems programmer looking to go deep on hardware, a researcher building performance-critical computational infrastructure, or a developer ready to move beyond surface-level GPU programming, this series is your definitive technical reference from first principles to expert-level mastery.