Read more
This practical guide shows how to analyse, manipulate or simulate scientific or other numerical data using the power of modern GPUs to greatly increase the speed of calculations. Aimed at researchers and graduate students, it contains numerous real-world examples in clear uncluttered C++ code. All example code is available online.
List of contents
1. Introduction to GPU kernels and hardware; 2. Thinking and coding in parallel; 3. Warps and cooperative groups; 4. Parallel stencils; 5. Textures; 6. Monte Carlo applications; 7. Concurrency using CUDA streams and events; 8. Application to PET scanners; 9. Scaling up; 10. Tools for profiling and debugging; 11. Tensor cores; A. A brief history of CUDA; B. Atomic operations; C. The NVCC complier; D. AVX and the Intel complier; E. Number formats; F. CUDA documentation and libraries; G. The CX header files; H. AI and Python; I. Topics in C++; Index.
About the author
Richard Ansorge is Emeritus University Senior Lecturer at the Cavendish Laboratory, University of Cambridge and Emeritus Tutor and Fellow at Fitzwilliam College, Cambridge. He is the author of over 170 peer-reviewed publications and co-author of the book The Physics and Mathematics of MRI (2016).
Summary
This practical guide shows how to analyse, manipulate or simulate scientific or other numerical data using the power of modern GPUs to greatly increase the speed of calculations. Aimed at researchers and graduate students, it contains numerous real-world examples in clear uncluttered C++ code. All example code is available online.
