Causality analysis and fault ascription in component-based systems - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Theoretical Computer Science Year : 2020

Causality analysis and fault ascription in component-based systems

Abstract

This article introduces a general framework for fault ascription, which consists in identifying, within a multi-component system, the components whose faulty behavior has caused the failure of said system. Our framework uses configuration structures as a general semantical model to handle truly concurrent executions, partial and distributed observations in a uniform way. As a first contribution, and in contrast with most of the current literature on counterfactual analysis which relies heavily on a set of toy examples, we first define a set of expected formal properties for counterfactual builders, i.e. operators that build counterfactual executions. We then show that causality analyses that satisfy our requirements meet a set of elementary soundness and completeness properties. Finally we present a concrete causality analysis meeting all our requirements, and we show that it behaves well under refinement. We present several examples illustrating various phenomena such as causal over-determination or observational determinism, and we discuss the relationship of our approach with Halpern and Pearl's actual causality analysis. This article is the published version of https://hal.inria.fr/hal-02161534.
Fichier principal
Vignette du fichier
S0304397520303510.pdf (822.43 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02927216 , version 1 (22-08-2022)

Licence

Attribution - NonCommercial

Identifiers

Cite

Gregor Gössler, Jean-Bernard Stefani. Causality analysis and fault ascription in component-based systems. Theoretical Computer Science, 2020, 837, pp.158-180. ⟨10.1016/j.tcs.2020.06.010⟩. ⟨hal-02927216⟩
97 View
44 Download

Altmetric

Share

Gmail Facebook X LinkedIn More