Biologically-inspired massively-parallel architectures- Computing beyond a million processors

Steve Furber; Andrew Brown

doi:10.1109/ACSD.2009.17

Biologically-inspired massively-parallel architectures- Computing beyond a million processors

Steve Furber, Andrew Brown

Advanced Processor Technology

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

The SpiNNaker project aims to develop parallel computer systems with more than a million embedded processors. The goal of the project is to support large-scale simulations of systems of spiking neurons in biological real time, an application that is highly parallel but also places very high loads on the communication infrastructure due to the very high connectivity of biological neurons. The scale of the machine requires fault-tolerance and power-efficiency to influence the design throughout, and the development has resulted in innovation at every level of design, including a self-timed inter-chip communication system that is resistant to glitch-induced deadlock and 'emergency' hardware packet re-routing around failed inter-chip links, through to run-time support for functional migration and real-time fault mitigation. © 2009 IEEE.

Original language	English
Title of host publication	Proceedings - International Conference on Application of Concurrency to System Design, ACSD\|Proc. Int. Conf. Appl. Concurrency Syst. Des. ACSD
Place of Publication	USA
Publisher	IEEE
Pages	3-12
Number of pages	9
ISBN (Print)	9780769536972
DOIs	https://doi.org/10.1109/ACSD.2009.17
Publication status	Published - 2009
Event	ACSD 2009 - 9th International Conference on Application of Concurrency to System Design - Augsburg Duration: 1 Jul 2009 → …

Conference

Conference	ACSD 2009 - 9th International Conference on Application of Concurrency to System Design
City	Augsburg
Period	1/07/09 → …

Keywords

biology computing , microprocessor chips , neural nets , parallel architectures

Access to Document

10.1109/ACSD.2009.17

Cite this

@inproceedings{867ac88730e042239a801c07c30d812b,

title = "Biologically-inspired massively-parallel architectures- Computing beyond a million processors",

abstract = "The SpiNNaker project aims to develop parallel computer systems with more than a million embedded processors. The goal of the project is to support large-scale simulations of systems of spiking neurons in biological real time, an application that is highly parallel but also places very high loads on the communication infrastructure due to the very high connectivity of biological neurons. The scale of the machine requires fault-tolerance and power-efficiency to influence the design throughout, and the development has resulted in innovation at every level of design, including a self-timed inter-chip communication system that is resistant to glitch-induced deadlock and 'emergency' hardware packet re-routing around failed inter-chip links, through to run-time support for functional migration and real-time fault mitigation. {\textcopyright} 2009 IEEE.",

keywords = "biology computing , microprocessor chips , neural nets , parallel architectures",

author = "Steve Furber and Andrew Brown",

year = "2009",

doi = "10.1109/ACSD.2009.17",

language = "English",

isbn = "9780769536972",

pages = "3--12",

booktitle = "Proceedings - International Conference on Application of Concurrency to System Design, ACSD|Proc. Int. Conf. Appl. Concurrency Syst. Des. ACSD",

publisher = "IEEE",

address = "United States",

note = "ACSD 2009 - 9th International Conference on Application of Concurrency to System Design ; Conference date: 01-07-2009",

}

Furber, S & Brown, A 2009, Biologically-inspired massively-parallel architectures- Computing beyond a million processors. in Proceedings - International Conference on Application of Concurrency to System Design, ACSD|Proc. Int. Conf. Appl. Concurrency Syst. Des. ACSD. IEEE, USA, pp. 3-12, ACSD 2009 - 9th International Conference on Application of Concurrency to System Design, Augsburg, 1/07/09. https://doi.org/10.1109/ACSD.2009.17

Biologically-inspired massively-parallel architectures- Computing beyond a million processors. / Furber, Steve; Brown, Andrew.
Proceedings - International Conference on Application of Concurrency to System Design, ACSD|Proc. Int. Conf. Appl. Concurrency Syst. Des. ACSD. USA: IEEE, 2009. p. 3-12.

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Biologically-inspired massively-parallel architectures- Computing beyond a million processors

AU - Furber, Steve

AU - Brown, Andrew

PY - 2009

Y1 - 2009

N2 - The SpiNNaker project aims to develop parallel computer systems with more than a million embedded processors. The goal of the project is to support large-scale simulations of systems of spiking neurons in biological real time, an application that is highly parallel but also places very high loads on the communication infrastructure due to the very high connectivity of biological neurons. The scale of the machine requires fault-tolerance and power-efficiency to influence the design throughout, and the development has resulted in innovation at every level of design, including a self-timed inter-chip communication system that is resistant to glitch-induced deadlock and 'emergency' hardware packet re-routing around failed inter-chip links, through to run-time support for functional migration and real-time fault mitigation. © 2009 IEEE.

AB - The SpiNNaker project aims to develop parallel computer systems with more than a million embedded processors. The goal of the project is to support large-scale simulations of systems of spiking neurons in biological real time, an application that is highly parallel but also places very high loads on the communication infrastructure due to the very high connectivity of biological neurons. The scale of the machine requires fault-tolerance and power-efficiency to influence the design throughout, and the development has resulted in innovation at every level of design, including a self-timed inter-chip communication system that is resistant to glitch-induced deadlock and 'emergency' hardware packet re-routing around failed inter-chip links, through to run-time support for functional migration and real-time fault mitigation. © 2009 IEEE.

KW - biology computing , microprocessor chips , neural nets , parallel architectures

U2 - 10.1109/ACSD.2009.17

DO - 10.1109/ACSD.2009.17

M3 - Conference contribution

SN - 9780769536972

SP - 3

EP - 12

BT - Proceedings - International Conference on Application of Concurrency to System Design, ACSD|Proc. Int. Conf. Appl. Concurrency Syst. Des. ACSD

PB - IEEE

CY - USA

T2 - ACSD 2009 - 9th International Conference on Application of Concurrency to System Design

Y2 - 1 July 2009

ER -

Biologically-inspired massively-parallel architectures- Computing beyond a million processors

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this