Automatic Data Layout Transformation for Heterogeneous Many-Core Systems

Abstract : Applying appropriate data structures is critical to attain superior performance in heterogeneous many-core systems. A heterogeneous many-core system is comprised of a host for control flow management, and a device for massive parallel data processing. However, the host and device require different types of data structures. The host prefers Array-of-Structures (AoS) to ease the programming, while the device requires Structure-of-Arrays (SoA) for efficient data accesses. The conflicted preferences cost excessive effort for programmers to transform the data structures between two parts. The separately designed kernels with different coding styles also cause difficulty in maintaining programs. This paper addresses this issue by proposing a fully automated data layout transformation framework. Programmers can maintain the code in AoS style on the host, while the data layout is converted into SoA when being transferred to the device. The proposed framework streamlines the design flow and demonstrates up to 177% performance improvement.
Type de document :
Communication dans un congrès
Ching-Hsien Hsu; Xuanhua Shi; Valentina Salapura. 11th IFIP International Conference on Network and Parallel Computing (NPC), Sep 2014, Ilan, Taiwan. Springer, Lecture Notes in Computer Science, LNCS-8707, pp.208-219, 2014, Network and Parallel Computing. 〈10.1007/978-3-662-44917-2_18〉
Liste complète des métadonnées

Littérature citée [9 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01403085
Contributeur : Hal Ifip <>
Soumis le : vendredi 25 novembre 2016 - 14:30:43
Dernière modification le : vendredi 1 décembre 2017 - 01:10:05
Document(s) archivé(s) le : lundi 20 mars 2017 - 19:03:42

Fichier

978-3-662-44917-2_18_Chapter.p...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Citation

Ying-Yu Tseng, Yu-Hao Huang, Bo-Cheng Lai, Jiun-Liang Lin. Automatic Data Layout Transformation for Heterogeneous Many-Core Systems. Ching-Hsien Hsu; Xuanhua Shi; Valentina Salapura. 11th IFIP International Conference on Network and Parallel Computing (NPC), Sep 2014, Ilan, Taiwan. Springer, Lecture Notes in Computer Science, LNCS-8707, pp.208-219, 2014, Network and Parallel Computing. 〈10.1007/978-3-662-44917-2_18〉. 〈hal-01403085〉

Partager

Métriques

Consultations de la notice

44

Téléchargements de fichiers

34