This roadmap describes how to combine parallel programming techniques from MPI and OpenMP when writing applications for multi-node HPC systems, such as Frontera and Stampede3, and explains the motivation for doing so.


After you complete this workshop, you should be able to:

  • List the principles that motivate a blend of shared- and distributed-memory parallel programming styles for multi-node HPC architectures
  • Demonstrate compiling and running advanced programs that combine MPI and OpenMP parallelization techniques on multi-node HPC systems
  • You should have some basic working knowledge of MPI and OpenMP, at the level of the Cornell Virtual Workshop topics on Message Passing Interface (MPI) and OpenMP.
  • You should also be familiar with common Linux shell commands, in bash, csh, or similar; otherwise, try working through the Linux topic first.

System requirements include:

  • A TACC or ACCESS account to log on to a TACC HPC system
  • A computation allocation for a TACC HPC system
©   Cornell University  |  Center for Advanced Computing  |  Copyright Statement  |  Inclusivity Statement