Node.js Addons for High Performance Numeric Computing
summary
Builds and benchmarks C/C++/Fortran bindings that actually link to real numeric libraries, showing how to compile Fortran with node-gyp and NAN (or migrate to N-API), how C wrappers call Fortran BLAS, and where costs hide in typed-array unwrapping. A reference Fortran BLAS add-on becomes nearly 6× faster for large arrays, hardware‑optimized MKL/Accelerate/OpenBLAS yield up to ~20×, but Windows Fortran/toolchain hassles force shipping fallbacks.
AI-written summary · about
chapters (5)
related talks
25mIntroduction to native extensions
31mBuilding Modern Native Add-ons for Node.js in 2021
55mNode.js N-API for Rust
01h 00mThe Blaze High Performance Math Library
39mNext Generation N-API: A Hands-on Workshop
21mModern C++ Addons for Node.js
34mWorker Threads vs C ++ Addons (part 2)
50mBuilding native Node.js addons in Reason