Accéder directement au contenu Accéder directement à la navigation
Article dans une revue

Performance Improvement of Stencil Computations for Multi-core Architectures based on Machine Learning

Abstract : Stencil computations are the basis to solve many problems related to Partial Differential Equations (PDEs). Obtaining the best performance with such numerical kernels is a major issue as many critical parameters (architectural features, compiler flags, memory policies, multithreading strategies) must be finely tuned. In this context, auto-tuning methods have been extensively used to improve the overall performance. However, the complexity of current architectures and the large number of optimizations to consider reduce the efficiency of this approach. This paper focuses on the use of Machine Learning to predict the performance of stencil kernels on multi-core architectures. Low-level hardware counters (e.g. cache-misses and TLB misses) on a limited number of executions are used to build our predictive model. We have considered two different kernels (7-point Jacobi and seismic wave modelling) to demonstrate the effectiveness of our approach. Our results show that performance can be predicted and that the best input configuration for stencil problems can be obtained by simulations of hardware counters and performance measurements.
Type de document :
Article dans une revue
Liste complète des métadonnées

https://hal-brgm.archives-ouvertes.fr/hal-01849636
Contributeur : Myriam Chergui <>
Soumis le : jeudi 26 juillet 2018 - 13:52:37
Dernière modification le : lundi 4 mai 2020 - 17:00:04

Lien texte intégral

Identifiants

Collections

Citation

Víctor Martínez, Fabrice Dupros, Márcio Castro, Philippe Navaux. Performance Improvement of Stencil Computations for Multi-core Architectures based on Machine Learning. Procedia Computer Science, Elsevier, 2017, 108, pp.305 - 314. ⟨10.1016/j.procs.2017.05.164⟩. ⟨hal-01849636⟩

Partager

Métriques

Consultations de la notice

64