Pruners: Providing reproducibility for uncovering non-deterministic errors in runs on supercomputers. (September 2019)

• Record Type: Journal Article
• Authors: Sato, Kento; Laguna, Ignacio; Lee, Gregory L; Schulz, Martin; Chambreau, Christopher M; Atzeni, Simone; Bentley, Michael; Gopalakrishnan, Ganesh; Rakamaric, Zvonimir; Sawaya, Geof; Protze, Joachim; Ahn, Dong H
• Other Names: Mascagni Michael guest-editor.
• Is Part Of: International journal of high performance computing applications. Volume 33:Number 5(2019)
• Page Start: 777
• Page End: 783
• Publication Date: 2019-09
• Subjects: Debugging -- testing -- reproducibility -- non-determinism -- MPI -- OpenMP; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342019834621
• Languages: English
• ISSNs: 1094-3420

Abstract

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.