MPI on TACC Systems
Christopher Cameron, Peter Vaillancourt, CAC Staff (original)
Cornell Center for Advanced Computing
Revisions: 5/2022, 3/2019, 6/2017, 2/2001 (original)
This topic is intended to guide researchers getting started with MPI on high performance computing systems, especially on Stampede2 and Frontera.
Objectives
After you complete this topic, you should be able to:
- Demonstrate compiling and running an MPI program on TACC's Frontera or Stampede2 systems.
- Describe important considerations when using nodes to parallelize code on Stampede2
- Explain why MPI programs cannot be run directly on TACC login nodes
Prerequisites
- A working knowledge of general programming concepts
- Ability to program in a high-level language such as Fortran, C, or C++
- A basic familiarity with parallel programming concepts