Parallel Scientific Computing - May 2011

This was an intensive week-long workshop on Scientific Parallel Programming, held at SciNet on the University of Toronto St. George campus.

Goal

The goal of the workshop was to enable young researchers already experienced in scientific computing to leave with the knowledge necessary to begin writing the parallel codes needed for their research. The workshop consisted of a mix of lectures and immediate feedback on practical assignments, designed to ensure that students leave with significant experience in both OpenMP and MPI, two of the standards for parallel computing today. GPGPU Programming with Cuda was also briefly introduced.

Slides (PDF)

1. Intro to Class and to Parallel Programming
2. SciNet Resources
3. C Review
4. Introduction to OpenMP
5. Introduction to MPI
6. Hydrodynamics
7. Version Control
8. Parallel Debugging
9. N-body Part 1 (openmp) Part 2 (mpi)
10. I/O Best practices on the parallel file system
11. CUDA GPU programming

Videos (to come)

1. Intro to Class and to Parallel Programming
   
2. SciNet Resources
   
3. C Review
   
4. Introduction to OpenMP - part 1part 2part 3part 4part 5part 6
   
5. Introduction to MPI Part 1Part 2
   
6. Hydrodynamics
   
7. HPC Best Practices and Version Control
   
8. Parallel Debugging
   
9. N-Body - part 1part 2
   
10. I/O on a Parallel File System
    
11. Intro to GPGPU Programming

Source Code

• Tar-ball of the source code used in class