Computer Science 263 - Parallel Computing Systems

This course takes a systems perspective on understanding parallel computing and processing. This course investigates both the hardware devices (CPU, GPU) that power modern parallel architectures as well as the layers of the software stack that are used to interface with these components. The course studies parallel computing systems through multiple lenses and paradigms, e.g. shared and distributed memory computing and GPU computing. And the course also uses a variety of programming models, languages, and frameworks: e.g. MPI, OpenMP, CUDA, Kokkos, etc, for harnessing that parallelism. Finally, the work touches on foundational theoretical aspects including parallel algorithmic design and analysis, and touches on other more advanced aspects like reproducibility, fault tolerance, parallel input/output, etc. Prerequisites: CSC 231. Enrollment limited to 30.

[CE] CSC 231