C++ Compute Efficiency: Algorithm & Arithmetic Optimizations (Advanced C++ Programming Book 11)

C++ Compute Efficiency is a book for C and C++ engineers looking to speed up their code.

Highlights :
- Low-level compute
- Memory access optimizations
- High-level algorithm speedups
- Compile-time metaprogramming.
- Data structure optimizations

Table of Contents: Part I: Arithmetic Optimizations

1. Expression Optimizations

2. Multiplication Optimizations

3. Fused Multiply-Add (FMA)

4. Division & Remainder

5. Approximate Arithmetic

6. Mathematical Approximations

7. Mathematical Algebra

8. Polynomial Approximation

9. Assembly & Immediates

10. ASCII/EBCDIC Tricks

11. Overflow, Saturation, Clamping

12. Timing & Benchmarking

Part II: Bitwise Arithmetic

13. Bitwise Operators

14. Bitwise Intrinsics

15. Bit Fiddling Magic

16. Floating-Point Bit Formats

17. Floating-Point Bit Magic

18. Block Floating Point

19. Fixed-Point Numbers

20. Advanced Number Systems

Part III: Compute Parallelism

21. AVX Intrinsics

22. AVX SIMD Vectorization

23. SWAR Integer Parallelism

24. Compute-Unit Parallelism

25. Instruction-Level Parallelism

Part IV: Control Flow Optimizations

26. Loop Unrolling

27. Loop Transformations

28. Short-Circuiting

29. Switch Multiway Branches

30. Goto Considered Efficient

31. Computed Goto Jump Tables

32. Branchless Introduction

33. Branchless Techniques

34. Branchless Select & Conditionals

35. Branchless Pitfalls

Part V: Function Call Optimizations

36. Inline Functions

37. Manual Inlining & Preprocessor Macros

38. Preprocessor Bugs

39. Code Bloat

40. Recursion Flattening

41. Closures, Lambdas and Functors

Part VI: Algorithm Optimizations

42. String Optimizations

43. Approximate Algorithms

44. Algorithm Speedups

45. Unrolled Binary Search

46. Vector Dot Product

47. Vector Approximations

48. Vector Algorithms

49. MatMul/GEMM

50. Bloom Filters

51. Deferred Computation

52. Prefilled Fixed-Size Data Structures

53. Runtime Auxiliary Precomputation

54. Data Compression

Part VII: Compile-Time Metaprogramming

55. Compile-Time Optimizations

56. Compiler Hints

57. Zero Runtime Cost Operations

58. Metaprogramming Precomputation

59. Lookup Table Precomputation

60. Old-Style C/C++ Source Code Generation

61. Text File Embedding

62. Data Structure Unrolling

63. Perfect Hashing Overview

64. Perfect Hash Metaprogramming

65. Old-Style Perfect Hashing in C

Part VIII: Memory Cache Optimizations

66. Prefetching

67. Page Prefetching

68. Pointer Arithmetic

69. Memory Reduction Optimizations

70. Contiguous Memory Blocks

71. Cache Warming

72. Cache Locality

73. Endianness & Byte Ordering

74. False Sharing

75. Intercepting Memory Primitives

76. Smart Pointers

Part VIII: LLM Compute Optimizations

77. Tensors

78. Low-Bit Quantization

79. Zero-Multiplication Models

80. Bitshift Quantization

81. Logarithmic Models

Appendices

Appendix A: Long List of Low Latency Techniques

Appendix B. 500 LLM Inference Optimization

Appendix C. 200 CUDA C++ Optimization Techniques

Appendix D: C++ Slug Catalog