Return this item by mail or in store within 90 days for a full refund.
Eligible for registries and wish lists
About this item
Highlights
Get the eBook free when you register your print book at Manning.
About the Author: Elliot Arledge created the 12-hour CUDA course and the 6-hour LLM from Scratch course for FreeCodeCamp, and consults on deep learning performance.
408 Pages
Computers, Data Science
Description
Book Synopsis
Get the eBook free when you register your print book at Manning."A fundamental guide for CUDA and modern GPU architecture and their application in the field of AI." --Bryce Adelstein Lelbach, NVIDIA The difference between a model that infers in milliseconds instead of seconds, or one that trains in days instead of weeks, is in how effectively you use your hardware. To squeeze every last teraflop out of an NVIDIA GPU, you need to write your own CUDA kernels, and this book will get you there! This book is a friendly hands-on tour of the modern GPU programming stack, written for the latest NVIDIA hardware, and grounded in fundamentals that will outlast any single generation of chips. In 408 fully-illustrated, code-rich pages, author Elliot Arledge guides you from your first kernel all the way to Flash Attention, tensor cores, and multi-GPU distributed training. GPU programming is complicated, so this book builds your skills one layer at a time using a six-level kernel optimization stack. You'll start by writing a vector addition kernel, learning how grids, blocks, and threads map onto real hardware. From there, you'll implement the naive versions of the kernels every deep learning model depends on: GEMM, softmax, 1D and 2D convolution, max pooling, and matrix transpose. Then you'll build and train a full neural network in six progressively faster implementations and dig into cuBLAS, with backpropagation derived and coded from scratch. Throughout this book, your focus will stay on real optimization work. You'll port a transformer to CUDA inference and attack each kernel in turn. Your GEMV decoder gets coalesced access, block-level reduction, and vectorization. Softmax becomes online, warp-shuffled, and fused. GEMM progresses through thread remapping, shared memory tiling, 1D and 2D blocktiling, and vectorized loads. Every optimization is profiled, and every profile is explained. The final chapters take you to the frontier. You'll program tensor cores using the WMMA and WGMMA instruction sets for Hopper hardware and see how manual tiling gets replaced by hardware-managed data movement. You'll implement Flash Attention with tiling and online softmax, quantize models, and explore scaling up to 16 GPUs with tensor and pipeline parallelism over NCCL. By the end, you'll have both a portfolio of more than 50 working kernels and the mental models to write new ones for hardware that doesn't exist yet. What's inside - CUDA and GPU programming from the ground up - Tensor and pipeline parallelism across multiple GPUs - PyTorch C++ extensions for shipping custom kernels About the reader For software and AI engineers comfortable with C/C++. No prior CUDA experience required. About the authorElliot Arledge is a GPU performance engineer and educator. Every kernel in this book comes from his workflow of writing, profiling, and verifying CUDA on both consumer and datacenter hardware. He created freeCodeCamp's 12-hour CUDA course and consults on deep learning performance.
About the Author
Elliot Arledge created the 12-hour CUDA course and the 6-hour LLM from Scratch course for FreeCodeCamp, and consults on deep learning performance.
Dimensions (Overall): 9.25 Inches (H) x 7.38 Inches (W)
Weight: .99 Pounds
Suggested Age: 22 Years and Up
Number of Pages: 408
Genre: Computers
Sub-Genre: Data Science
Publisher: Manning Publications
Theme: Machine Learning
Format: Paperback
Author: Elliot Arledge
Language: English
Street Date: December 1, 2026
TCIN: 1011604952
UPC: 9781633434899
Item Number (DPCI): 247-02-1143
Origin: Made in the USA or Imported
If the item details aren’t accurate or complete, we want to know about it.
Shipping details
Estimated ship dimensions: 1 inches length x 7.38 inches width x 9.25 inches height
Estimated ship weight: 0.99 pounds
We regret that this item cannot be shipped to PO Boxes.
This item cannot be shipped to the following locations: American Samoa (see also separate entry under AS), Guam (see also separate entry under GU), Northern Mariana Islands, Puerto Rico (see also separate entry under PR), United States Minor Outlying Islands, Virgin Islands, U.S., APO/FPO, Alaska, Hawaii
Return details
This item can be returned to any Target store or Target.com.
This item must be returned within 90 days of the date it was purchased in store, delivered to the guest, delivered by a Shipt shopper, or picked up by the guest.