Books

We will work in pure Rust, with no C++ shim or Python wrapper. We will build one small program that grows to cover the whole book. It begins as a loop that any of us could have written on a bad day. After gaining a kernel, it learns to stop waiting, grows streams and pinned memory, and ends as a pipeline fed by parallel producers through bounded channels. The device is busy almost every millisecond.

A local model returns the expected answer. How do we know which layers ran on the GPU? In Local AI on RTX Spark Laptops, we investigate questions like this through 251 hands-on WSL2 exercises, following the evidence from model-loader logs down to individual CUDA kernels.

Explore how modern GPU virtualization brings near-native performance to Linux virtual machines. From hardware and KVM/QEMU internals to virtio-nvgpu and NVIDIA's shared GPU architecture, this book takes you inside the stack with practical configurations, source code analysis and deployment patterns for real-world systems.

What I love about Dask is that it won't make us start from scratch. It doesn't ask us to learn a new language, adopt a new way of thinking, or rewrite five years of careful analysis into something completely different. It just takes the tools we've already got and quietly teaches them to work in parallel. The dataframe remains a dataframe. The array is still an array. The difference is that the work now spreads across eight cores, or eight hundred, and we barely had to change how we think.

Go beyond writing kernel modules and learn how real Linux GPU drivers are built. Starting with GPU architecture and the graphics stack, this hands-on guide walks you through creating a complete DRM driver, from a minimal module to memory management, command submission, display support and error recovery.

Unlock NVIDIA GPU performance from Rust. This practical guide takes you from GPU architecture and memory hierarchies to writing and optimizing real CUDA Rust kernels. Explore both official Rust CUDA approaches with runnable examples and learn how to build fast, production-ready GPU code without giving up Rust’s safety and clarity.

Cloud GPU rental costs more than compute time. It can also introduce recurring expenses, external dependencies, and additional questions about where your prompts, datasets, and model artifacts are processed. A local machine learning workstation gives you direct control over the hardware, software stack, and data workflows. The Local AI Stack is a technical engineering reference for building a Windows-based machine learning workstation with Linux environments, NVIDIA GPU acceleration, and virtualization. Across 140 pages, it follows the stack from firmware configuration and Windows virtualization features to CUDA-enabled containers, model inference, storage optimization, and system recovery.

An Advanced Engineering Reference for GPU Architecture, Low-Level Optimization, and High-Performance ComputingModern GPU programming requires more than exposing parallel work to thousands of threads. It demands an understanding of the relationship between hardware architecture, instruction execution, memory movement, synchronization, and the software systems that coordinate computation across multiple GPUs. Mastering NVIDIA CUDA: Expert GPU Programming is a 100-chapter technical reference for experienced developers, high-performance computing engineers, researchers, and AI systems architects.

Discover Mojo, a language that pairs Python-friendly syntax with serious systems-level speed. This hands-on guide takes you from your first program to optimized CPU and GPU workloads, with practical techniques for debugging, profiling and building production-ready software for high-performance computing and AI.

Build serious AI systems on NVIDIA DGX Spark with a practical guide that goes far beyond setup. Learn to deploy LLMs, create RAG pipelines, orchestrate AI agents, and optimize performance for real production workloads. Whether you are experimenting or scaling enterprise AI, this book shows you how to get there.

We'll be writing kernels in Rust and seeing how they match up with their C++ equivalents. If you feed a GPU properly, you'll see bandwidth quadruple from just one changed subscript. It'll get you working with thousands of threads, pricing a financial option based on an exact formula, and shipping an inference pipeline that outruns its Python equivalent for reasons you can name.

Unlock the full power of modern GPU programming with Mastering the Vulkan API. This practical guide helps intermediate C++ developers build high-performance graphics and compute applications through clear explanations, hands-on examples, and production-inspired best practices.

Learn CUDA programming from the ground up by understanding how GPUs really execute code. This book shows you how to write efficient kernels, optimize performance, and build high-performance applications for AI, scientific computing, image processing, and other demanding workloads on modern NVIDIA GPUs, including Blackwell.