Benchmarking Parallel Performance MPI Packages

This is my application test result of debian GSOC 2024 project Benchmarking Parallel Performance MPI Packages.

My Environment

Google Cloud Engine Debian 12 with 4 CPU core.

How to run

You have to be root to run this shell script.

sh iterate_benchmark.sh

How does it work

It will iterate over all the free BLAS implementation in Debian package, and use autopkgtest to run the fenicsx-performance-tests with different BLAS libraries respectively.

What result should it present

There should be a directory under your $HOME name fenics−benchmarking−rs, and inside it will have different directory which name by their used implementation of blas, which content all the logs of benchmarking.