Skip to Main content Skip to Navigation

Infrastructure pour la gestion générique et optimisée des traces d'exécution pour les systèmes embarqués.

Abstract : Validation is a critical aspect of system development. This is especially true in the context of embedded systems which come with various technical and physical constraints. As embedded systems have enormously grown in complexity in recent years, formal method validation is too complex or too costly to apply. Validation must therefore use post-conception methods, the major method being trace analysis. Indeed, execution traces capture many details about systems' execution and provide a means for advanced system analysis. However, trace analysis faces two major challenges. On one hand, trace analysis has to deal and support huge execution traces. On the other hand, analysis methods should be able to retrieve relevant metrics from the low-level information traces contain. This thesis has been done as part of the SoC-TRACE projet and presents three contributions. Our first contribution is a definition of a generic execution trace format that expresses semantics. Our second contribution is a workflow-based infrastructure for generic and automatic trace analysis. This infrastructure addresses the problem of huge traces management using streaming mechanisms. It allows modular and configurable analyses, as well as automatic analyses execution. Our third contribution concerns generic performance analysis for Linux systems. This contribution defines the means for trace recording, as well as an analysis workflow producing unified performance profiles. We validate our contributions with use cases given by STMicroelectronics and traces obtained from benchmark executions. Our trace format with semantics allowed us to automatically bring out execution problems. Using streaming mechanisms, we have been able to analyze traces that can reach several hundreds of gigabytes. Our generic analysis method for systems let us to automatically highlight, without any prior knowledge, internal behavior of benchmark programs. Our generic solutions point out a similar execution behavior of benchmarks on different machines and architectures, and showed their impact on the execution.
Complete list of metadatas
Contributor : Vania Marangozova-Martin <>
Submitted on : Monday, March 20, 2017 - 9:14:28 AM
Last modification on : Friday, July 17, 2020 - 11:12:03 AM
Long-term archiving on: : Wednesday, June 21, 2017 - 12:33:26 PM


  • HAL Id : tel-01492474, version 1


Alexis Martin. Infrastructure pour la gestion générique et optimisée des traces d'exécution pour les systèmes embarqués. . Calcul parallèle, distribué et partagé [cs.DC]. Université Grenoble Alpes, 2017. Français. ⟨tel-01492474v1⟩



Record views


Files downloads