The SpiNNaker project

Steve B. Furber; Francesco Galluppi; Steve Temple; Luis A. Plana

doi:10.1109/JPROC.2014.2304638

The SpiNNaker project

Steve B. Furber, Francesco Galluppi, Steve Temple, Luis A. Plana

Advanced Processor Technology

Research output: Contribution to journal › Article › peer-review

Abstract

The spiking neural network architecture (SpiNNaker) project aims to deliver a massively parallel million-core computer whose interconnect architecture is inspired by the connectivity characteristics of the mammalian brain, and which is suited to the modeling of large-scale spiking neural networks in biological real time. Specifically, the interconnect allows the transmission of a very large number of very small data packets, each conveying explicitly the source, and implicitly the time, of a single neural action potential or 'spike.' In this paper, we review the current state of the project, which has already delivered systems with up to 2500 processors, and present the real-time event-driven programming model that supports flexible access to the resources of the machine and has enabled its use by a wide range of collaborators around the world. © 2014 IEEE.

Original language	English
Article number	6750072
Pages (from-to)	652-665
Number of pages	13
Journal	Institute of Electrical and Electronics Engineers. Proceedings
Volume	102
Issue number	5
DOIs	https://doi.org/10.1109/JPROC.2014.2304638
Publication status	Published - 27 Feb 2014

Keywords

Brain modeling
multicast algorithms
multiprocessor interconnection networks
neural network hardware
parallel programming

Access to Document

10.1109/JPROC.2014.2304638Licence: CC BY

Cite this

@article{9fed179f612a405b8801b67ef74bc737,

title = "The SpiNNaker project",

abstract = "The spiking neural network architecture (SpiNNaker) project aims to deliver a massively parallel million-core computer whose interconnect architecture is inspired by the connectivity characteristics of the mammalian brain, and which is suited to the modeling of large-scale spiking neural networks in biological real time. Specifically, the interconnect allows the transmission of a very large number of very small data packets, each conveying explicitly the source, and implicitly the time, of a single neural action potential or 'spike.' In this paper, we review the current state of the project, which has already delivered systems with up to 2500 processors, and present the real-time event-driven programming model that supports flexible access to the resources of the machine and has enabled its use by a wide range of collaborators around the world. {\textcopyright} 2014 IEEE.",

keywords = "Brain modeling, multicast algorithms, multiprocessor interconnection networks, neural network hardware, parallel programming",

author = "Furber, {Steve B.} and Francesco Galluppi and Steve Temple and Plana, {Luis A.}",

year = "2014",

month = feb,

day = "27",

doi = "10.1109/JPROC.2014.2304638",

language = "English",

volume = "102",

pages = "652--665",

journal = "Institute of Electrical and Electronics Engineers. Proceedings ",

issn = "0018-9219",

publisher = "IEEE",

number = "5",

}

TY - JOUR

T1 - The SpiNNaker project

AU - Furber, Steve B.

AU - Galluppi, Francesco

AU - Temple, Steve

AU - Plana, Luis A.

PY - 2014/2/27

Y1 - 2014/2/27

N2 - The spiking neural network architecture (SpiNNaker) project aims to deliver a massively parallel million-core computer whose interconnect architecture is inspired by the connectivity characteristics of the mammalian brain, and which is suited to the modeling of large-scale spiking neural networks in biological real time. Specifically, the interconnect allows the transmission of a very large number of very small data packets, each conveying explicitly the source, and implicitly the time, of a single neural action potential or 'spike.' In this paper, we review the current state of the project, which has already delivered systems with up to 2500 processors, and present the real-time event-driven programming model that supports flexible access to the resources of the machine and has enabled its use by a wide range of collaborators around the world. © 2014 IEEE.

AB - The spiking neural network architecture (SpiNNaker) project aims to deliver a massively parallel million-core computer whose interconnect architecture is inspired by the connectivity characteristics of the mammalian brain, and which is suited to the modeling of large-scale spiking neural networks in biological real time. Specifically, the interconnect allows the transmission of a very large number of very small data packets, each conveying explicitly the source, and implicitly the time, of a single neural action potential or 'spike.' In this paper, we review the current state of the project, which has already delivered systems with up to 2500 processors, and present the real-time event-driven programming model that supports flexible access to the resources of the machine and has enabled its use by a wide range of collaborators around the world. © 2014 IEEE.

KW - Brain modeling

KW - multicast algorithms

KW - multiprocessor interconnection networks

KW - neural network hardware

KW - parallel programming

U2 - 10.1109/JPROC.2014.2304638

DO - 10.1109/JPROC.2014.2304638

M3 - Article

SN - 0018-9219

VL - 102

SP - 652

EP - 665

JO - Institute of Electrical and Electronics Engineers. Proceedings

JF - Institute of Electrical and Electronics Engineers. Proceedings

IS - 5

M1 - 6750072

ER -

The SpiNNaker project

Abstract

Keywords

Access to Document

Fingerprint

Cite this