An efficient scheduling algorithm for dataflow architecture using loop-pipelining

doi:10.1016/j.ins.2020.09.029

	An efficient scheduling algorithm for dataflow architecture using loop-pipelining
	Li, Yi 1,3; Wu, Meng 1; Ye, Xiaochun 1,2; Li, Wenming 1; Xue, Rui 1,3; Wang, Da 1; Zhang, Hao 1; Fan, Dongrui 1,3
刊名	INFORMATION SCIENCES
	2021-02-08
卷号	547 页码:1136-1153
关键词	Dataflow architecture Instruction scheduling Multicast Sharing path Loop optimization
ISSN号	0020-0255
DOI	10.1016/j.ins.2020.09.029
英文摘要	Dataflow architecture has native advantages in achieving high instruction parallelism and power efficiency for today's emerging applications such as high performance computing and deep neural network. In dataflow computing, the execution of instructions is driven by data, so the data transfer efficiency of the network on chip (NoC) is a key factor affecting performance. However, the NoC performance degrades due to the increasing use of multi-cast communications in many applications. The existing dataflow architecture instruction scheduling algorithms do not optimize multicast communication between the instruction and its successor instructions, so the routing paths of many multicast packets have forks which cause bandwidth waste and potential network congestion. We propose a sharing path awareness (SPA) algorithm to optimize multicast communication in the dataflow architecture. The algorithm shares the routing paths from the instruction to its child node to reduce the NoC bandwidth waste through the instruction scheduler. For applications using software iteration, we further extend the loop optimization to the SPA algorithm to sufficiently exploit instruction-level parallelism. Compared with the state-of-the-art algorithm, we show that the SPA algorithm achieves 20.21% average performance improvement and 15.11% energy consumption reduction for our experimental workloads. (C) 2020 Elsevier Inc. All rights reserved.
资助项目	National Key Research and Development Program[2018YFB1003501] ; National Natural Science Foundation of China[61732018] ; National Natural Science Foundation of China[61872335] ; National Natural Science Foundation of China[61802367] ; National Natural Science Foundation of China[61672499] ; Austrian-Chinese Cooperative RD Project (FFG)[171111KYSB20170032] ; Austrian-Chinese Cooperative RD Project (CAS)[171111KYSB20170032] ; Strategic Priority Research Program of Chinese Academy of Sciences[XDC05000000] ; Open Project Program of the State Key Laboratory of Mathematical Engineering and Advanced Computing[2019A07]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER SCIENCE INC
WOS记录号	WOS:000590678700008
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/16525]
专题	中国科学院计算技术研究所
通讯作者	Ye, Xiaochun
作者单位	1.Chinese Acad Sci, State Key Lab Comp Architecture, ICT, Beijing, Peoples R China 2.State Key Lab Math Engn & Adv Comp, Wuxi, Jiangsu, Peoples R China 3.UCAS, Sch Comp & Control Engn, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Li, Yi,Wu, Meng,Ye, Xiaochun,et al. An efficient scheduling algorithm for dataflow architecture using loop-pipelining[J]. INFORMATION SCIENCES,2021,547:1136-1153.
APA	Li, Yi.,Wu, Meng.,Ye, Xiaochun.,Li, Wenming.,Xue, Rui.,...&Fan, Dongrui.(2021).An efficient scheduling algorithm for dataflow architecture using loop-pipelining.INFORMATION SCIENCES,547,1136-1153.
MLA	Li, Yi,et al."An efficient scheduling algorithm for dataflow architecture using loop-pipelining".INFORMATION SCIENCES 547(2021):1136-1153.