ODAJIMA, Tetsuya; BOKU, Taisuke; SATO, Mitsuhisa; HANAWA, Toshihiro; KODAMA, Yuetsu; NAMYST, Raymond; THIBAULT, Samuel; AUMAGE, Olivier

doi:10.1007/978-3-319-03889-6_7

Métadonnées

Licence d’utilisation du document

ODAJIMA, Tetsuya
Graduate School of Systems and Information Engineering [Tsukuba]

BOKU, Taisuke
Graduate School of Systems and Information Engineering [Tsukuba]
Center for Computational Sciences [Tsukuba] [CCS]

SATO, Mitsuhisa
Center for Computational Sciences [Tsukuba] [CCS]
Graduate School of Systems and Information Engineering [Tsukuba]

Langue

Communication dans un congrès

Ce document a été publié dans

The 2013 International Symposium on Advances of Distributed and Parallel Computing (ADPC 2013), 2013-12-18, Vietri sul Mare. 2013-12-19

Résumé en anglais

On the work sharing among GPUs and CPU cores on GPU equipped clusters, it is a critical issue to keep load balance among these heterogeneous computing resources. We have been developing a runtime system for this problem on PGAS language named XcalableMP- dev/StarPU [1]. Through the development, we found the necessity of adaptive load balancing for GPU/CPU work sharing to achieve the best performance for various application codes. In this paper, we enhance our language system XcalableMP-dev/StarPU to add a new feature which can control the task size to be assigned to these heterogeneous resources dynamically during application execution. As a result of performance evaluation on several benchmarks, we confirmed the proposed feature correctly works and the performance with heterogeneous work sharing provides up to about 40% higher performance than GPU-only utilization even for relatively small size of problems.< Réduire

Métadonnées

Partager cette publication !

Licence d’utilisation du document

Adaptive Task Size Control on High Level Programming for GPU/CPU Work Sharing

Langue

Ce document a été publié dans

Résumé en anglais

URI

DOI

Origine

Unités de recherche