📖 Book · hpc · advanced

🚀 High-Performance Computing in C++

The four levers of parallelism — SIMD, OpenMP, cache blocking, parallel STL, GPU & MPI — runnable.

HPC as a fight against the memory wall: arithmetic intensity and the roofline, then the four levers of parallelism with runnable code — SIMD auto-vectorization, OpenMP reductions, cache blocking/tiling, and C++17 parallel STL — plus the GPU (CUDA/SIMT) and MPI (distributed-memory) models for scaling beyond one core.

6 chapters▶ runnable code✦ in-depth animation
▶ See it animated:SIMD & Auto-Vectorization →OpenMP: shared-memory parallelism →Parallel STL (std::execution) →Cache blocking & loop tiling →

What this book covers

  1. 01HPC is a fight against the memory wall
  2. 02SIMD: many elements per instruction
  3. 03OpenMP: fill every core
  4. 04Cache blocking: reuse data while it's hot
  5. 05Parallel STL: parallelism without the plumbing
  6. 06Scaling out: GPU and MPI

Read it in full — runnable C++ + step-through animations

Every chapter of High-Performance Computing in C++ ships complete, editable code you compile and run in the browser, plus the same in-depth animations as the lessons.

Start free →Already a member? Log in →