This workshop is offered by the National Center for Supercomputing Applications. Each exercise is comprised of a serial code and the OpenMP-parallelized version of it. Makefiles are available for compilation with GCC, as well as batch job scripts with specific configurations to run the code on different systems.
1. hello
This exercise is the classic Hello, World! code. It serves as an introduction to how threads can execute tasks independently and asynchronously.
2. stats
The STATS application calculates simples statistical problems of a large collection of real numbers extracted from a uniform probability distribution and organized into an array. It serves as an introduction to the concept of Task Parallelism.
3. saxpy
The SAXPY application performs a very common operation in Linear Algebra-related packages. It serves as an introduction to the concept of Data Parallelism.
This repository is intended to be used during live training sessions. More exercises can be found in the Parallelization GitHub repo. If you would to contribute with a different exercise, I would be delighted to hear from you!