UCL-ARC / householder

A native Rust library for advanced Linear Algebra routines
https://docs.rs/householder/latest/householder/
Apache License 2.0
21 stars 4 forks source link

Efficient Matrix-Matrix Products #2

Open tbetcke opened 2 years ago

tbetcke commented 2 years ago

Any efficient linear algebra library requires fast matrix-matrix products. The Rust ndarray project uses the matrixmultiply library https://github.com/bluss/matrixmultiply, which uses the same ideas on which the BLIS project is based. This seems to be a very good start.

In this issue we want to track the use of matrixmultiply and its performance compared to Openblas on different architectures.

jedbrown commented 1 year ago

Is this issue complete in the sense that householder uses matrixmultiply?

tbetcke commented 1 year ago

I leave it open on purpose because of performance evaluation. I'd like to have good comparison metrics against Openblas on different architectures.

jedbrown commented 1 year ago

Yep, or use the BLIS benchmarking infrastructure to test matrixmultiply. https://github.com/flame/blis/blob/master/docs/Performance.md

Or this could be done entirely in Rust, sweeping over problem sizes for each factorization and scalar, but I'd encourage plotting it normalized to match the BLIS figures.