Unicorn: A Multicore Neuromorphic Processor with Flexible Fan-In and Unconstrained Fan-Out for Neurons

被引：5

作者：

Yang, Zhijie ^{[1
]}

Wang, Lei ^{[1
]}

Wang, Yao ^{[1
]}

Peng, Linghui ^{[1
]}

Chen, Xiaofan ^{[1
]}

Xiao, Xun ^{[1
]}

Wang, Yaohua ^{[1
]}

Xu, Weixia ^{[1
]}

机构：

[1] Natl Univ Def Technol, Coll Comp Sci & Technol, Changsha, Hunan, Peoples R China

来源：

PROCEEDINGS OF THE 59TH ACM/IEEE DESIGN AUTOMATION CONFERENCE, DAC 2022 | 2022年

基金：

国家重点研发计划;

关键词：

multicore architecture; neuromorphic processor; spiking neural network; dynamic vision sensor; hardware accelerator; LARGE-SCALE MODEL;

D O I：

10.1145/3489517.3530563

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Neuromorphic processor is popular due to its high energy efficiency for spatio-temporal applications. However, when running the spiking neural network (SNN) topologies with the ever-growing scale, existing neuromorphic architectures face challenges due to their restrictions on neuron fan-in and fan-out. This paper proposes Unicorn, a multicore neuromorphic processor with a spike train sliding multicasting mechanism (STSM) and neuron merging mechanism (NMM) to support unconstrained fan-out and flexible fan-in of neurons. Unicorn supports 36K neurons and 45M synapses and thus supports a variety of neuromorphic applications. The peak performance and energy efficiency of Unicorn reach 36TSOPS and 424GSOPS/W respectively. Experimental results show that Unicorn can achieve 2x-5.5x energy reduction over the state-of-the-art neuromorphic processor when running an SNN with a relatively large fan-out and fan-in.

引用

页码：943 / 948

页数：6

共 50 条

[11] Ultra-Low Crosstalk Fused Taper Type Fan-in/Fan-out Devices for Multicore Fibers
Gan, Lin
Zhou, Jiajun
Shen, Li
Guo, Xiancong
Wang, Yanlin
Yang, Chen
Tong, Weijun
Xia, Li
Fu, Songnian
Tang, Ming
Liu, Deming
2019 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION (OFC), 2019,
[12] FAST DC COUPLED ANALOGUE FAN-IN AND FAN-OUT MODULES
HERBST, LJ
NUCLEAR INSTRUMENTS & METHODS, 1969, 70 (02): : 185 - &
[13] ANALYSIS OF WEIGHTED FAN-OUT FAN-IN VOLUME HOLOGRAPHIC OPTICAL INTERCONNECTIONS
ASTHANA, P
NORDIN, GP
TANGUAY, AR
JENKINS, BK
APPLIED OPTICS, 1993, 32 (08) : 1441 - 1469
[14] Parameter-based refactoring and the relationship with fan-in/fan-out coupling
Murgia, Alessandro
Tonelli, Roberto
Concas, Giulio
Marchesi, Michele
Counsell, Steve
JOURNAL OF OBJECT TECHNOLOGY, 2012, 11 (02):
[15] Impact of gate fan-in and fan-out limits on optoelectronic digital circuits
Ji, LH
Heuring, VP
APPLIED OPTICS, 1997, 36 (17): : 3927 - 3940
[16] NOR OPTICAL LOGIC GATE ARRAY - FAN-IN AND FAN-OUT MEASUREMENTS
GUALOUS, H
KOSTER, A
PASCAL, D
LAVAL, S
INTERNATIONAL JOURNAL OF OPTOELECTRONICS, 1994, 9 (06): : 471 - 476
[17] BOUNDED FAN-IN BOUNDED FAN-OUT UNIFORM DECOMPOSITIONS OF SYNCHRONOUS SEQUENTIAL MACHINES
WEINER, P
HOPCROFT, JE
PROCEEDINGS OF THE INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS, 1968, 56 (07): : 1219 - &
[18] A Longitudinal Study of Fan-In and Fan-Out Coupling in Open-Source Systems
Mubarak, Asma
Counsell, Steve
Hierons, Robert M.
INTERNATIONAL JOURNAL OF INFORMATION SYSTEM MODELING AND DESIGN, 2011, 2 (04) : 1 - 26
[19] Adaptively Secure KP-ABE For Circuits With Fan-In n And Fan-Out 1
Sun, Keshuo
Gao, Haiying
COMPUTER JOURNAL, 2022, 66 (10): : 2554 - 2573
[20] Lens-integrated Fan-in/Fan-out Device for Multi-core Fiber
Shimakawa, Osamu
Arao, Hajime
Sano, Tomomi
2016 21ST OPTOELECTRONICS AND COMMUNICATIONS CONFERENCE (OECC) HELD JOINTLY WITH 2016 INTERNATIONAL CONFERENCE ON PHOTONICS IN SWITCHING (PS), 2016,

← 1 2 3 4 5 →