Scientific Libraries

Who likes to reinvent the wheel? While coding yourself numerical methods undeniably gives you great skills, let’s face it, we often don’t have the time nor the expertise to write state-of-the-art algorithms. Here is a collection of links that point to these very good computing libraries. Some of them may save you days, weeks, months of hard developments. I will continuously update this page so add it in your bookmarks and come back from time to time ;-)

  • PETSC : Huge numerical parallel library (C, C++, Fortran, python…)
  • GSL : Gnu Scientific Library, simple to use library for basic numerical programs
  • OpenMPI : MPI implementation
  • Using MPI 2nd Edition : Really good book to learn MPI programming
  • Cours MPI de l’IDRIS : Nice MPI tuto, in french
  • Zoltan : Nice library for domain decomposition problems, dynamic load balancing …