Enabling In-Network Floating-Point Arithmetic for Efficient Computation Offloading

doi:10.1109/TPDS.2022.3208425

	Enabling In-Network Floating-Point Arithmetic for Efficient Computation Offloading
	Cui, Penglai 1,2; Pan, Heng 1,3; Li, Zhenyu 1,3; Zhang, Penghao 1,2; Miao, Tianhao 1,2; Zhou, Jianer 4; Guan, Hongtao 1; Xie, Gaogang 5
	2022-12-01
发表期刊	IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
ISSN	1045-9219
卷号	33 期号:12 页码:4918-4934
摘要	Programmable switches are recently used for accelerating data-intensive distributed applications. Some computational tasks, traditionally performed on servers in data centers, are offloaded into the network on programmable switches. These tasks may require the support of on-the-fly floating-point operations. Unfortunately, programmable switches are restricted to simple integer arithmetic operations. Existing systems circumvent this restriction by converting floats to integers or relying on local CPUs of switches, incurring extra processing delayed and accuracy loss. To address this gap, we propose NetFC, a table-lookup method to achieve on-the-fly in-network floating-point arithmetic operations nearly without accuracy loss. Specifically, NetFC utilizes logarithm projection and transformation to convert the original huge table enumerating all operands and results into several much smaller tables that can fit into the data plane of programmable switches. To cope with the table inflation problem on 32-bit floats, we also propose an approximation method that further breaks the large tables into smaller ones. In addition, NetFC leverages two optimizations to improve accuracy and reduce on-chip memory consumption. We use both synthetic and real-life datasets to evaluate NetFC. The experimental results show that the average accuracy of NetFC is above 99.9% with only 448KB memory consumption for 16-bit floats and 99.1% with 496KB memory consumption for 32-bit floats. Furthermore, we integrate NetFC into two distributed applications and two in-network telemetry systems to show its effectiveness in further improving the performance.
关键词	Open area test sites Arithmetic Memory management Task analysis Training Standards Servers In-network computation computation offloading floating-point operation
DOI	10.1109/TPDS.2022.3208425
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China[2020YFB1805600] ; National Natural Science Foundation of China[U20A20180] ; National Natural Science Foundation of China[62002344] ; Beijing Natural Science Foundation[JQ20024]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:000864178200003
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/19793
专题	中国科学院计算技术研究所期刊论文
通讯作者	Li, Zhenyu
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 3.Purple Mt Labs, Nanjing 211111, Peoples R China 4.Southern Univ Sci & Technol, Shenzhen 518055, Peoples R China 5.Chinese Acad Sci, Comp Network Informat Ctr, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Cui, Penglai,Pan, Heng,Li, Zhenyu,et al. Enabling In-Network Floating-Point Arithmetic for Efficient Computation Offloading[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,2022,33(12):4918-4934.
APA	Cui, Penglai.,Pan, Heng.,Li, Zhenyu.,Zhang, Penghao.,Miao, Tianhao.,...&Xie, Gaogang.(2022).Enabling In-Network Floating-Point Arithmetic for Efficient Computation Offloading.IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,33(12),4918-4934.
MLA	Cui, Penglai,et al."Enabling In-Network Floating-Point Arithmetic for Efficient Computation Offloading".IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 33.12(2022):4918-4934.