Programming Massively Parallel Processors: A Hands-on Approach shows both students and professionals alike the basic concepts of parallel programming and GPU architecture. Concise, intuitive, and practical, it is based on years of road-testing in the authors' own parallel computing courses. Various techniques for constructing and optimizing parallel programs are explored in detail, while case studies demonstrate the development process, which begins with computational thinking and ends with effective and efficient parallel programs. The new edition includes updated coverage of CUDA, including the newer libraries such as CuDNN. New chapters on frequently used parallel patterns have been added, and case studies have been updated to reflect current industry practices.
1. Chapter 1 Introduction - p. 1 2. Chapter 2 Heterogeneous data parallel computing - p. 23 3. Chapter 3 Multidimensional grids and data - p. 47 4. Chapter 4 Compute architecture and scheduling - p. 69 5. Chapter 5 Memory architecture and data locality - p. 93 6. Chapter 6 Performance considerations - p. 123 7. Chapter 7 Convolution - p. 151 8. Chapter 8 Stencil - p. 173 9. Chapter 9 Parallel histogram - p. 191 10. Chapter 10 Reduction - p. 211 11. Chapter 11 Prefix sum (scan) - p. 235 12. Chapter 12 Merge - p. 263 13. Chapter 13 Sorting - p. 293 14. Chapter 14 Sparse matrix computation - p. 311 15. Chapter 15 Graph traversal - p. 331 16. Chapter 16 Deep learning - p. 355 17. Chapter 17 Iterative magnetic resonance imaging reconstruction - p. 391 18. Chapter 18 Electrostatic potential map - p. 415 19. Chapter 19 Parallel programming and computational thinking - p. 433 20. Chapter 20 Programming a heterogeneous computing cluster - p. 449 21. Chapter 21 CUDA dynamic parallelism - p. 475 22. Chapter 22 Advanced practices and future evolution - p. 499 23. Chapter 23 Conclusion and outlook - p. 515 24. Appendix A Numerical considerations - p. 519 25. Index - p. 537
Programming Massively Parallel Processors: A Hands-on Approach shows both students and professionals alike the basic concepts of parallel programming and GPU architecture. Concise, intuitive, and practical, it is based on years of road-testing in the authors' own parallel computing courses. Various techniques for constructing and optimizing parallel programs are explored in detail, while case studies demonstrate the development process, which begins with computational thinking and ends with effective and efficient parallel programs. The new edition includes updated coverage of CUDA, including the newer libraries such as CuDNN. New chapters on frequently used parallel patterns have been added, and case studies have been updated to reflect current industry practices.
1. Chapter 1 Introduction - p. 1 2. Chapter 2 Heterogeneous data parallel computing - p. 23 3. Chapter 3 Multidimensional grids and data - p. 47 4. Chapter 4 Compute architecture and scheduling - p. 69 5. Chapter 5 Memory architecture and data locality - p. 93 6. Chapter 6 Performance considerations - p. 123 7. Chapter 7 Convolution - p. 151 8. Chapter 8 Stencil - p. 173 9. Chapter 9 Parallel histogram - p. 191 10. Chapter 10 Reduction - p. 211 11. Chapter 11 Prefix sum (scan) - p. 235 12. Chapter 12 Merge - p. 263 13. Chapter 13 Sorting - p. 293 14. Chapter 14 Sparse matrix computation - p. 311 15. Chapter 15 Graph traversal - p. 331 16. Chapter 16 Deep learning - p. 355 17. Chapter 17 Iterative magnetic resonance imaging reconstruction - p. 391 18. Chapter 18 Electrostatic potential map - p. 415 19. Chapter 19 Parallel programming and computational thinking - p. 433 20. Chapter 20 Programming a heterogeneous computing cluster - p. 449 21. Chapter 21 CUDA dynamic parallelism - p. 475 22. Chapter 22 Advanced practices and future evolution - p. 499 23. Chapter 23 Conclusion and outlook - p. 515 24. Appendix A Numerical considerations - p. 519 25. Index - p. 537
