GPU Programming in MATLAB

GPU programming in MATLAB is intended for scientists, engineers, or students who develop or maintain applications in MATLAB and would like to accelerate their codes using GPU programming without losing the many benefits of MATLAB. The book starts with coverage of the Parallel Computing Toolbox and other MATLAB toolboxes for GPU computing, which allow applications to be ported straightforwardly onto GPUs without extensive knowledge of GPU programming. The next part covers built-in, GPU-enabled features of MATLAB, including options to leverage GPUs across multicore or different computer systems. Finally, advanced material includes CUDA code in MATLAB and optimizing existing GPU applications. Throughout the book, examples and source codes illustrate every concept so that readers can immediately apply them to their own development.

• Provides in-depth, comprehensive coverage of GPUs with MATLAB, including the parallel computing toolbox and built-in features for other MATLAB toolboxes
• Explains how to accelerate computationally heavy applications in MATLAB without the need to re-write them in another language
• Presents case studies illustrating key concepts across multiple fields
• Includes source code, sample datasets, and lecture slides

Scientists working in MATLAB who wish to leverage GPUs; high performance computing engineers wishing to incorporate MATLAB; students studying these topics

• Dedication
• About the Authors
• Foreword
• Preface
• Chapter 1: Introduction
  • Abstract
  • 1.1 Parallel Programming
  • 1.2 GPU Programming
  • 1.3 CUDA Architecture
  • 1.4 Why GPU Programming in MATLAB? When to Use GPU Programming?
  • 1.5 Our Approach: Organization of the Book
  • 1.6 Chapter Review
• Chapter 2: Getting started
  • Abstract
  • Chapter Objectives
  • 2.1 Hardware Requirements
  • 2.2 Software Requirements
  • 2.2.1 NVIDIA CUDA Toolkit
  • 2.3 Chapter Review
• Chapter 3: Parallel Computing Toolbox
  • Abstract
  • 3.1 Product Description and Objectives
  • 3.2 Parallel for-Loops (parfor)
  • 3.3 Single Program Multiple Data (spmd)
  • 3.4 Distributed and Codistributed Arrays
  • 3.5 Interactive Parallel Development (pmode)
  • 3.6 GPU Computing
  • 3.7 Clusters and Job Scheduling
  • 3.8 Chapter Review
• Chapter 4: Introduction to GPU programming in MATLAB
  • Abstract
  • 4.1 GPU Programming Features in MATLAB
  • 4.2 GPU Arrays
  • 4.3 Built-in MATLAB Functions for GPUs
  • 4.4 Element-Wise MATLAB Code on GPUs
  • 4.5 Chapter Review
• Chapter 5: GPU programming on MATLAB toolboxes
  • Abstract
  • 5.1 Communications System Toolbox
  • 5.2 Image Processing Toolbox
  • 5.3 Neural Network Toolbox
  • 5.4 Phased Array System Toolbox
  • 5.5 Signal Processing Toolbox
  • 5.6 Statistics and Machine Learning Toolbox
  • 5.7 Chapter Review
• Chapter 6: Multiple GPUs
  • Abstract
  • 6.1 Identify and Run Code on a Specific GPU Device
  • 6.2 Examples Using Multiple GPUs
  • 6.3 Chapter Review
• Chapter 7: Run CUDA or PTX code
  • Abstract
  • 7.1 A Brief Introduction to CUDA C
  • 7.2 Steps to Run CUDA or PTX Code on a GPU Through MATLAB
  • 7.3 Example: Vector Addition
  • 7.4 Example: Matrix Multiplication
  • 7.5 Chapter Review
• Chapter 8: MATLAB MEX functions containing CUDA code
  • Abstract
  • 8.1 A Brief Introduction to MATLAB MEX Files
  • 8.2 Steps to Run MATLAB MEX Functions on GPU
  • 8.3 Example: Vector Addition
  • 8.4 Example: Matrix Multiplication
  • 8.5 Chapter Review
• Chapter 9: CUDA-accelerated libraries
  • Abstract
  • 9.1 Introduction
  • 9.2 cuBLAS
  • 9.3 cuFFT
  • 9.4 cuRAND
  • 9.5 cuSOLVER
  • 9.6 cuSPARSE
  • 9.7 NPP
  • 9.8 Thrust
  • 9.9 Chapter Review
• Chapter 10: Profiling code and improving GPU performance
  • Abstract
  • 10.1 MATLAB Profiling
  • 10.2 CUDA Profiling
  • 10.3 Best Practices for Improving GPU Performance
  • 10.4 Chapter Review
• References
• List of Examples
• Index