GPU-SAM: Leveraging multi-GPU split-and-merge execution for system-wide real-time support
详细信息 查看全文
- 作者：Wookhyun HanAuthor Vitae ; Hoon Sung ChwaAuthor Vitae ; Hwidong BaeAuthor Vitae ; Hyosu KimAuthor Vitae ; Insik Shin ; insik.shin@cs.kaist.ac.kr" class="auth_mail" title="E-mail the corresponding author ; insik.shin@gmail.com" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Real-time systems ; Multi-GPU ; GPGPU
- 刊名：Journal of Systems and Software
- 出版年：2016
- 出版时间：July 2016
- 年：2016
- 卷：117
- 期：Complete
- 页码：1-14
- 全文大小：981 K
文摘
- •
We examine benefits and costs of split-and-merge execution on multi-GPU systems.
- •
The split-and-merge execution can improve schedulability on real-time systems.
- •
We model schedulability analysis for split-and-merge execution.
- •
We propose an algorithm called GPA, to decide the number of GPUs to be used.
- •
We demonstrate through evaluations that GPA can improve system-wide schedulability.