Adaptive Line Size Cache for Irregular References on Cell Multicore Processor

Abstract : Software cache promises to achieve programmability on Cell processor. However, irregular references couldn't achieve a considerable performance improvement since the cache line is always set to a specific size. In this paper, we propose an adaptive cache line prefetching strategy which continuously adjusts cache line size during application execution. Therefore, the transferred data is decreased significantly. Moreover, a corresponding software cache - adaptive line size cache is designed. It introduces a hybrid Tag Entry Arrays, with each mapping to a different line size. It's a hierarchical design in that the misshandler is not invoked immediately when an address is a miss in the short line Tag Entry Array. Instead, the long line Tag Entry Array is checked first, which significantly increases the hit rate. Evaluations indicate that improvement due to the adaptive cache line strategy translates into 3.29 to 5.73 speedups compared to the traditional software cache approach.
Type de document :
Communication dans un congrès
Chen Ding; Zhiyuan Shao; Ran Zheng. IFIP International Conference on Network and Parallel Computing (NPC), Sep 2010, Zhengzhou, China. Springer, Lecture Notes in Computer Science, LNCS-6289, pp.314-328, 2010, Network and Parallel Computing. 〈10.1007/978-3-642-15672-4_27〉
Liste complète des métadonnées

Littérature citée [19 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01054976
Contributeur : Hal Ifip <>
Soumis le : lundi 11 août 2014 - 09:13:15
Dernière modification le : vendredi 11 août 2017 - 17:44:29
Document(s) archivé(s) le : mercredi 26 novembre 2014 - 21:36:42

Fichier

Adaptive_Line_Size_Cache_for_I...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Citation

Qian Cao, Chongchong Zhao, Junxiu Chen, Yunxing Zhang, Yi Chen. Adaptive Line Size Cache for Irregular References on Cell Multicore Processor. Chen Ding; Zhiyuan Shao; Ran Zheng. IFIP International Conference on Network and Parallel Computing (NPC), Sep 2010, Zhengzhou, China. Springer, Lecture Notes in Computer Science, LNCS-6289, pp.314-328, 2010, Network and Parallel Computing. 〈10.1007/978-3-642-15672-4_27〉. 〈hal-01054976〉

Partager

Métriques

Consultations de la notice

54

Téléchargements de fichiers

64