Register by November 23, 2018 to a seminar on the topic of OpenMP!
The second seminar in a series Information Technology is offering focuses on showing how OpenMP works and how to write code that uses it. Jarno van der Kolk will show how to set up OpenMP and share strategies to get the best performance. Information how to compile your code, how to program using threads, and what the limitations are will be discussed.
OpenMP is a library that enables parallel processing for programs written in C, C++, or Fortran. Even existing programs can be relatively easy changed from serial to parallel.
OpenMP offers a relatively simple way of letting your programs using multiple CPU cores to potentially gain a significant boost in performance. In an ideal scenario, using all cores on a quad-core CPU could cut your runtime by a factor of four.
The seminar will also touch on how to run this code on the computational resources provided by Compute Canada, where you can have access to systems with up to 64 CPU cores.
The code examples for this seminar are available on BitBucket.