Fast and Accurate Context-Aware Basic Block Timing Prediction using Transformers - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Fast and Accurate Context-Aware Basic Block Timing Prediction using Transformers

Résumé

This paper introduces ORXESTRA, a context-aware execution time prediction model based on Transformers XL, specifically designed to accurately estimate performance in embedded system applications. Unlike traditional machine learning models that often overlook contextual information, resulting in biased predictions for individual isolated basic blocks, ORXESTRA overcomes this limitation by incorporating execution context awareness. By doing so, ORXESTRA effectively accounts for the processor micro-architecture without explicitly modeling micro-architectural elements such as caches, pipelines, and branch predictors. Our evaluations demonstrate ORXESTRA's ability to provide precise timing estimations for different ARM targets (Cortex M4, M7, A53, and A72), surpassing existing machine learningbased approaches in both prediction accuracy and prediction speed.
Fichier principal
Vignette du fichier
3640537.3641572.pdf (741.19 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04406073 , version 1 (19-01-2024)
hal-04406073 , version 2 (18-03-2024)

Licence

Paternité

Identifiants

Citer

Abderaouf Nassim Amalou, Elisa Fromont, Isabelle Puaut. Fast and Accurate Context-Aware Basic Block Timing Prediction using Transformers. CC 2024 - ACM SIGPLAN 33rd International Conference on Compiler Construction, Gabriel Rodríguez, Mar 2024, Edimbourg, United Kingdom. pp.227-237, ⟨10.1145/3640537.3641572⟩. ⟨hal-04406073v2⟩
51 Consultations
66 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More