AMA: Asynchronous Management of Accelerators for Task-based Programming Models

Computational science has benefited in the last years from emerging accelerators that increase the performance of scientific simulations, but using these devices hinders the programming task. This paper presents AMA: a set of optimization techniques to efficiently manage multi-accelerator systems. A...

Full description

Saved in:
Bibliographic Details
Published in:Procedia computer science Vol. 51; pp. 130 - 139
Main Authors: Planas, Judit, Badia, Rosa M., Ayguade, Eduard, Labarta, Jesus
Format: Journal Article Publication
Language:English
Published: Elsevier B.V 2015
Elsevier
Subjects:
Accelerator management
Arquitectura de computadors
Asynchronous devices
Graphics processing units
Informàtica
Multi-GPU systems
Multiprocessadors
Multiprocessors
Parallel processing (Electronic computers)
Processament en paral·lel (Ordinadors)
Programming models
Àrees temàtiques de la UPC
Asynchronous devices
Accelerator management
Programming models
Multi-GPU systems
ISSN:1877-0509, 1877-0509
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Computational science has benefited in the last years from emerging accelerators that increase the performance of scientific simulations, but using these devices hinders the programming task. This paper presents AMA: a set of optimization techniques to efficiently manage multi-accelerator systems. AMA maximizes the overlap of computation and communication in a blocking-free way. Then, we can use such spare time to do other work while waiting for device operations. Implemented on top of a task-based framework, the experimental evaluation of AMA on a quad-GPU node shows that we reach the performance of a hand-tuned native CUDA code, with the advantage of fully hiding the device management. In addition, we obtain up to more than 2x performance speed-up with respect to the original framework implementation.
ISSN:1877-0509
1877-0509
DOI:10.1016/j.procs.2015.05.212