This is the repo that backs BYU's Scientific Computing Course, which is generated by GitHub Pages.
To pull and run the website locally:
git clone [email protected]:BYUHPC/sci-comp-course.git
cd sci-comp-course
# I had to do this, not sure if that's just a setup problem on my end
export GEM_HOME="$(ruby -e 'puts Gem.user_dir')"
export GEM_PATH="$GEM_HOME"
# Install Ruby stuff and run
bundle install
bundle exec jekyll serve"Shell and Slurm Practice" is meh; it'd be nice to have a find thrown in there (plus material on find), and make it more open-ended and amenable to combining sed/grep/awk.
Our Linux tutorial could be basically a drop-in replacement for a lot of lesson 2.
Once LLVM or GCC supports compiling DPC++ for GPUs, use that rather than nvc++ for the GPU phase. Looks like AdaptiveCpp might be a good option since it can compile for AMD GPUs as well.
Updating to C++23 will be nice, especially since we'll finally get std::mdspan. Once compilers finally support modules, all the code should probably be updated to use them since they're nicer than headers.