arrow
返回

Efficient algorithms for task mapping on heterogeneous CPU/GPU platforms for fast completion time

delete2021-03-01
delete15
PRE
AI
Z
Zexin Li
Y
Yuqun Zhang *
A
Ao Ding
H
Husheng Zhou
C
Cong Liu
DOI:10.1016/j.sysarc.2020.101936delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
In GPU-based embedded systems, the problem of computation and data mapping for multiple applications while minimizing the completion time is quite challenging due to large size of the policy space. To achieve fast competition time, a fine-grain mapping framework that explores a set of critical factors is needed for heterogeneous embedded systems. In this paper, we present a theoretical framework that yields a sub-optimal solution via three practical mapping algorithms with low time complexity. We evaluate such algorithms upon StarPU with a large set of popular benchmarks. Experimental results demonstrate that algorithms proposed by the original EMSOFT paper can achieve up to 30% faster completion time compared to state-of-the-art mapping techniques, and can perform consistently well across different workloads. We further extend such algorithms to minimize the completion time and enhance the runtime performance of complex heterogeneous applications under resource-limited infrastructure. We also extend the evaluation by deploying StarPU under multiple setups with an additional benchmark testing suite for simulating real-world runtime neural networks. Experimental results demonstrate that our extended algorithm can achieve much faster completion time (averagely 30% to 37% under multiple resource-constraint scenarios) compared to the state-of-the-art mapping techniques.
Keyword:
GPU
Heterogeneous scheduling
Data-size-based prediction
Neural network runtime acceleration
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Journal of Systems Architecture 封面图
Journal of Systems Architecture
IF:
4.1
论文数:
3.0K
被引数:
4.2K

机构

U
university of texas system
学者数:
18.5W
论文数: 15.6W
被引数: 210
引用论文

引用论文

err分享
err收藏
Lateral overgrowth mechanisms and microstructural characteristics of bulk-like GaN layers grown by sublimation method侧向过生长机制及由升华法生长的体材料GaN层的微观结构特征
err1999-02-01
err0
PREAI
errJie Wang; R. S. Qhalid Fareed; M. Hao; S. Mahanty; S. Tottori; Y. Ishikawa; T. Sugahara; Y. Morishima; K. Nishino; M. Osinski; S. Sakai
err分享
err收藏
Electron Temperature Gradient Scale at Collisionless Shocks
err2011-11-14
err0
errOAAI
errSteven J. Schwartz; Edmund Henley; Jeremy Mitchell; Vladimir Krasnoselskikh
err分享
err收藏
Adaptive suppression of power line interference in ultra-low field magnetic resonance imaging in an unshielded environment
err2018-01-01
err0
errOAAI
errXiaolei Huang; Hui Dong; Yang Qiu; Bo Li; Quan Tao; Yi Zhang; Hans-Joachim Krause; Andreas Offenhäusser; Xiaoming Xie
err分享
err收藏
学者 查看更多内容