The course’s goal is to confront students with the computing performance’s problem in a context where modern architectures are parallel (multi-core and GPU).

At the end of this course students are able to think their algorithms for the parallel paradigm and implement them on CPU (OpenMP) and on GPU (CUDA).

Parallel codes are implemented in C ++ using OpenMP and CUDA.

Course material

Course documents are available on Edunao to students registered for the course.