P runers : Providing reproducibility for uncovering non-deterministic errors in runs on supercomputers
Kento Sato, Ignacio Laguna, Gregory L. Lee, Martin Schulz, Christopher M. Chambreau, Simone Atzeni, Michael Bentley, Ganesh Lalitha Gopalakrishnan, Zvonimir Rakamarić, Geof Sawaya, Joachim Protze, Dong H. Ahn · The International Journal of High Performance Computing Applications · 2019
Large scientific simulations must be able to achieve the full-system potential of supercomputers. When they tap into high-performance features, however, a phenomenon known as non-determinism may be introduced in their program execution, which significantly hampers application development. Pruners is a new toolset to detect and remedy non-deterministic bugs and errors in large parallel applications. To show the capabilities of Pruners for large application development, we also demonstrate their early usage on real-world production applications.