I’ve been using Ada for scientific computing since some years ago, as I find it a really good language for that purpose. Having support for matrix operations in its standard library makes it really easy to develop that type of programs, as it reduces the need of managing third party dependencies (needing to use something like Conan and cmake for small optimization programs introduces way more overhead, so Ada feels like a better replacement for MATLAB). Additionally, Gnat library for managing variable units is quite nice for avoiding common mistakes that happen in scientific programs.
In my experience, using a typed language helps managing better memory consumption in those types of programs and Ada has a better performance than python. For that reason, I was quite surprised when I couldn’t find almost any project using it for that purpose (I was trying to find how to properly design an Ada program for that kind of applications, but Google kept suggesting me to use python and Jupiter notebooks hehehe).
Do you know if there’s any research programming their algorithms in Ada for running them in an HPC cluster? (The distributed annex and map reduce algorithm in Ada 202x seems a good starting point for that use, but I was trying to find how other projects use it)
