A three-phase workflow for general and expressive representations of nondeterminism in HPC applications. (November 2019)

• Record Type: Journal Article
• Authors: Chapp, Dylan; Rorabaugh, Danny; Sato, Kento; Ahn, Dong H; Taufer, Michela
• Other Names: Dongarra Jack guest-editor.; Tourancheau Bernard guest-editor.
• Is Part Of: International journal of high performance computing applications. Volume 33:Number 6(2019)
• Page Start: 1175
• Page End: 1184
• Publication Date: 2019-11
• Subjects: Nondeterminism -- event graph -- graph kernels -- reproducibility -- replicability -- transparency -- trace analysis -- graph similarity -- clustering; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342019868826
• Languages: English
• ISSNs: 1094-3420

Abstract

Nondeterminism is an increasingly entrenched property of high-performance computing (HPC) applications and has recently been shown to seriously hamper debugging and reproducibility efforts. Tools for addressing the nondeterministic debugging problem have emerged, but they do not provide methods for systematically cataloging the nondeterminism in a given application. We propose a three-phase workflow for representing executions of nondeterministic message passing interface programs as event graphs, quantifying their structural similarity with graph kernels, and applying machine learning techniques to investigate shared properties across applications. We present an empirical study comparing two graph kernels' suitability for this task and propose future uses of the methodology.