Measurement Based Execution Time Analysis of GPGPU Programs via SE+GA

Measurement Based Execution Time Analysis of GPGPU Programs via SE+GA
复制标题

通过 SE GA 对 GPGPU 程序进行基于测量的执行时间分析

DOI:
10.1109/dsd.2018.00021
复制
发表时间:
2018
期刊:
2018 21st Euromicro Conference on Digital System Design (DSD)
影响因子:
--
通讯作者:
Zebo Peng
Zebo Peng
中科院分区:
--
文献类型:
--
作者:
Adrian Horga;Sudipta Chattopadhyay;P. Eles;Zebo Peng

文献摘要

参考文献

被引文献

相似文献

了解执行时间对于嵌入式实时应用程序至关重要。最坏情况执行时间(WCET)是衡量嵌入式应用程序实时性的一个重要指标。对于复杂的执行平台,如图形处理单元(GPU),WCET的分析提出了很大的挑战,由于GPU的体系结构以及GPU程序语义的复杂特性。在本文中,我们提出了GDivAn,一个基于测量的WCET分析工具,用于任意GPU内核。GDivAn系统地结合了符号执行(SE)和遗传算法(GA)的优势,以保持分析过程的可扩展性和有效性。我们对几个开源GPU内核的评估揭示了GDivAn的效率。
Understanding the execution time is critical for embedded, real-time applications. Worst-case execution time (WCET) is an important metric to check the real-time constraints imposed on embedded applications. For complex execution platforms, such as graphics processing units (GPUs), analysis of WCET imposes great challenges due to the complex characteristics of GPU architecture as well as GPU program semantics. In this paper, we propose GDivAn, a measurement-based WCET analysis tool for arbitrary GPU kernels. GDivAn systematically combines the strength of symbolic execution (SE) and genetic algorithm (GA) to maintain both the scalability and the effectiveness of the analysis process. Our evaluation with several open-source GPU kernels reveals the efficiency of GDivAn.
使用混合分析估计 GPU 加速应用程序的 WCET
DOI: 10.1109/ecrts.2013.29
发表时间: 2013
期刊: --
影响因子: --
作者:
Betts A
通讯作者: Betts A