TAIL: Exploiting Temporal Asynchronous Execution for Efficient Spiking Neural Networks with Inter-Layer Parallelism

Spiking neural networks (SNNs) are an alternative computational paradigm to artificial neural networks (ANNs) that have attracted attention due to their event-driven execution mechanisms, enabling extremely low energy consumption. However, the existing SNN execution model, based on software simulati...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings - Design, Automation, and Test in Europe Conference and Exhibition s. 1 - 7
Hlavní autoři: Li, Haomin, Liu, Fangxin, Wang, Zongwu, Lyu, Dongxu, Huang, Shiyuan, Yang, Ning, Sun, Qi, Song, Zhuoran, Jiang, Li
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: EDAA 31.03.2025
Témata:
ISSN:1558-1101
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Spiking neural networks (SNNs) are an alternative computational paradigm to artificial neural networks (ANNs) that have attracted attention due to their event-driven execution mechanisms, enabling extremely low energy consumption. However, the existing SNN execution model, based on software simulation or synchronized hardware circuitry, is incompatible with the event-driven nature, thus resulting in poor performance and energy efficiency. The challenge arises from the fact that neuron computations across multiple time steps result in increased latency and energy consumption. To overcome this bottleneck and leverage the full potential of SNNs, we propose TAIL, a pioneering temporal asynchronous execution mechanism for SNNs driven by a comprehensive analysis of SNN computations. Additionally, we propose an efficient dataflow design to support SNN inference, enabling concurrent computation of various time steps across multiple layers for optimal Processing Element (PE) utilization. Our evaluations show that TAIL greatly improves the performance of SNN inference, achieving a 6.94× speedup and a 6.97× increase in energy efficiency on current SNN computing platforms.
AbstractList Spiking neural networks (SNNs) are an alternative computational paradigm to artificial neural networks (ANNs) that have attracted attention due to their event-driven execution mechanisms, enabling extremely low energy consumption. However, the existing SNN execution model, based on software simulation or synchronized hardware circuitry, is incompatible with the event-driven nature, thus resulting in poor performance and energy efficiency. The challenge arises from the fact that neuron computations across multiple time steps result in increased latency and energy consumption. To overcome this bottleneck and leverage the full potential of SNNs, we propose TAIL, a pioneering temporal asynchronous execution mechanism for SNNs driven by a comprehensive analysis of SNN computations. Additionally, we propose an efficient dataflow design to support SNN inference, enabling concurrent computation of various time steps across multiple layers for optimal Processing Element (PE) utilization. Our evaluations show that TAIL greatly improves the performance of SNN inference, achieving a 6.94× speedup and a 6.97× increase in energy efficiency on current SNN computing platforms.
Author Liu, Fangxin
Huang, Shiyuan
Sun, Qi
Song, Zhuoran
Jiang, Li
Lyu, Dongxu
Wang, Zongwu
Li, Haomin
Yang, Ning
Author_xml – sequence: 1
  givenname: Haomin
  surname: Li
  fullname: Li, Haomin
  email: haominli@sjtu.edu.cn
  organization: Shanghai Jiao Tong University
– sequence: 2
  givenname: Fangxin
  surname: Liu
  fullname: Liu, Fangxin
  email: liufangxin@sjtu.edu.cn
  organization: Shanghai Jiao Tong University
– sequence: 3
  givenname: Zongwu
  surname: Wang
  fullname: Wang, Zongwu
  organization: Shanghai Jiao Tong University
– sequence: 4
  givenname: Dongxu
  surname: Lyu
  fullname: Lyu, Dongxu
  organization: Shanghai Jiao Tong University
– sequence: 5
  givenname: Shiyuan
  surname: Huang
  fullname: Huang, Shiyuan
  organization: Shanghai Jiao Tong University
– sequence: 6
  givenname: Ning
  surname: Yang
  fullname: Yang, Ning
  organization: Shanghai Jiao Tong University
– sequence: 7
  givenname: Qi
  surname: Sun
  fullname: Sun, Qi
  organization: Zhejiang University
– sequence: 8
  givenname: Zhuoran
  surname: Song
  fullname: Song, Zhuoran
  organization: Shanghai Jiao Tong University
– sequence: 9
  givenname: Li
  surname: Jiang
  fullname: Jiang, Li
  email: ljiang_cs@sjtu.edu.cn
  organization: Shanghai Jiao Tong University
BookMark eNo1kE1OwzAYRA0CCVq4AQtfIMU_SWyzi0qASlFBIqwrx_lMTVM7ilOV3p5WwGoWb94sZoIufPCAEKZkxrii6v6xqMs8zZmcMcKyGSVKcaL4GZpwJVkuUkrEObqmWSYTSgm9QpMYvwghGWfqGsW6WFQPuPzuu-BG5z9xDds-DLrDRTx4sx6CD7t4LIDZjS54bMOAS2udceBH_N67zclawu7kLGHch2ET8d6Na7zwIwxJpQ8w4Dd95B10Lm5v0KXVXYTbv5yij6eynr8k1evzYl5UiaNCjklrqNI0bZRqpeQpUGWtbqFptFCN1YYbqZngjOTC8FYQ25ocZCMtESw3rOVTdPe76wBg1Q9uq4fD6v8h_gP9I2Bk
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.23919/DATE64628.2025.10993093
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
EISBN 3982674107
9783982674100
EISSN 1558-1101
EndPage 7
ExternalDocumentID 10993093
Genre orig-research
GrantInformation_xml – fundername: Natural Science Foundation of Shanghai
  grantid: 24ZR1433700
  funderid: 10.13039/100007219
– fundername: National Natural Science Foundation of China
  grantid: 62402311
  funderid: 10.13039/501100001809
GroupedDBID 6IE
6IF
6IH
6IK
6IL
6IN
AAJGR
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
FEDTE
IEGSK
IPLJI
KZ1
LMP
M43
OCL
RIE
RIL
ID FETCH-LOGICAL-i178t-dc19a14b99d8834e19ffadebba79bfac3c8a2732067c3d70fdc6e8b8f0726c2d3
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001506972600290&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 01:55:02 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i178t-dc19a14b99d8834e19ffadebba79bfac3c8a2732067c3d70fdc6e8b8f0726c2d3
PageCount 7
ParticipantIDs ieee_primary_10993093
PublicationCentury 2000
PublicationDate 2025-March-31
PublicationDateYYYYMMDD 2025-03-31
PublicationDate_xml – month: 03
  year: 2025
  text: 2025-March-31
  day: 31
PublicationDecade 2020
PublicationTitle Proceedings - Design, Automation, and Test in Europe Conference and Exhibition
PublicationTitleAbbrev DATE
PublicationYear 2025
Publisher EDAA
Publisher_xml – name: EDAA
SSID ssj0005329
Score 2.2869499
Snippet Spiking neural networks (SNNs) are an alternative computational paradigm to artificial neural networks (ANNs) that have attracted attention due to their...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Asynchronous Execution
Brain-inspired Computing
Computational modeling
Energy consumption
Energy efficiency
Hardware
Parallel processing
Performance gain
Software
Spiking neural networks
Synchronization
Tail
Title TAIL: Exploiting Temporal Asynchronous Execution for Efficient Spiking Neural Networks with Inter-Layer Parallelism
URI https://ieeexplore.ieee.org/document/10993093
WOSCitedRecordID wos001506972600290&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NT8IwGG6EeNALihi_04PXAV3Z2nojCtGEEBJnwo30M1mCgzAw8d_7tgPRgwdvy7JuS99ufd6P53kRugf_jFGZyIipJIkA34pIaTBItyt5SlOdGCdDswk2HvPpVEy2ZPXAhbHWhuIz2_aHIZdvFnrjQ2Udn8XxmbsaqjGWVmStfT0HjUVVqhNTQUTnqZ8NUs-8BCcwTtq7sb-6qIRNZNj45-NPUGtPx8OT743mFB3Yookau34MePt5NtHxD3HBM1Rm_ZfRAw5FdrkvbsZZJUM1x_3ys9BeFRfcfrjA6rD6MOBXPAiSEvAe-HWZ-zA69vIdMGZc1YuX2EducQgkRiMJgB1P5Mo3ZJnn5XsLvQ0H2eNztG2xEOWE8XVkNBGS9JQQhnPas0Q4J41VSjKhnNRUcwkAx2u8a2pY1xmdWq6467I41bGh56heLAp7gbCR8MNKFEmdIT0Ct9EArbQDBK9SDajiErX8lM6WlYrGbDebV3-cv0ZH3nAV_-8G1derjb1Fh_pjnZeru2D7L-8gstg
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NT8IwGG4UTdQLihi_7cHrYF330XojCoE4CYkz4Ub6mSzBQRiY-O9tOxA9ePC2LOvStN369H2f53kBuDfnswSziHkJjyLP4FvqcWEmxPcZiXEsIqmZKzaRDIdkPKajtVjdaWGUUo58plr20uXy5UysbKisbbM4NnO3C_aiMAz8Sq61ZXTggFZknQBTRNtPnawbW-2lOQYGUWvT-lcdFbeN9Or_7MAxaG4FeXD0vdWcgB1VNEB9U5EBrj_QBjj6YS94CsqsM0gfoKPZ5ZbeDLPKiGoKO-VnIawvrjn4mweUcOsPGgQLu85UwvQDvs5zG0iH1sDDtBlWjPES2tgtdKFEL2UGssMRW9iSLNO8fG-Ct143e-x76yILXo4SsvSkQJShkFMqCcGhQlRrJhXnLKFcM4EFYQbiWJd3gWXiayliRTjRfhLEIpD4DNSKWaHOAZTM_LIijmItUYjMa4QBV0IbDM9jYXDFBWjaIZ3MKx-NyWY0L_-4fwcO-tlLOkkHw-crcGgnsVIDXoPacrFSN2BffCzzcnHr1sEXZOy2Hw
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=proceeding&rft.title=Proceedings+-+Design%2C+Automation%2C+and+Test+in+Europe+Conference+and+Exhibition&rft.atitle=TAIL%3A+Exploiting+Temporal+Asynchronous+Execution+for+Efficient+Spiking+Neural+Networks+with+Inter-Layer+Parallelism&rft.au=Li%2C+Haomin&rft.au=Liu%2C+Fangxin&rft.au=Wang%2C+Zongwu&rft.au=Lyu%2C+Dongxu&rft.date=2025-03-31&rft.pub=EDAA&rft.eissn=1558-1101&rft.spage=1&rft.epage=7&rft_id=info:doi/10.23919%2FDATE64628.2025.10993093&rft.externalDocID=10993093