Computational Skills for Researchers: MPI/OpenMP
Instructors: Pascal Paschos, Senior Computational Specialist, Northwestern IT Research Computing Services and Alper Kinaci, Senior Computational Specialist, Northwestern IT Research Computing Services
OpenMP (Open Multi-Processing) and MPI (Message Passing Interface) are two standard approaches to parallelize codes. OpenMP provides parallelism within a multi-core node and MPI offers intra- and inter-node parallelization. This introductory workshop will focus on the fundamental concepts of parallel computing using these two methods. Topics will include distributed versus shared memory computing, constructing parallel regions, domain decomposition and avoiding race conditions. The workshop will include examples of using MPI with Python. This workshop assumes you're comfortable writing at least basic Python code. Please bring your own laptop; installation instructions will be provided ahead of the workshop.