it is not easy for a compiler to vectorize a pragmatic approach: write in a high level interpreted language that rhymes with modern CPUs, vector extensions, me

it is not easy for a compiler to vectorize a pragmatic approach: write in a high level interpreted language that rhymes with modern CPUs, vector extensions, memory bandwidth e.g. apl [0], bqn [1], k [2], kiwi [3] - vectors are dense (not boxed) - optimized internal representation (e.g. bitpacked bool vectors) - primitives act on vectors + use avx, neon if possible [0] https://www.dyalog.com https://www.dyalog.com [1] https://mlochbaum.github.io/BQN/ https://mlochbaum.github.io/BQN/ [2] https://kx.com https://kx.com [3] https://kiwilang.com https://kiwilang.com great article by marshall on BQN performance compared to C and how to think about it https://mlochbaum.github.io/BQN/implementation/versusc.html https://mlochbaum.github.io/BQN/implementation/versusc.html related: - columnar databases: kdb, duckdb, clickhouse - machine learning frameworks: pytorch, keras, jax, mlx