Maintaining real-time synchrony on SpiNNaker

Sergio Davies; Alexander D. Rast; Francesco Galluppi; Steve B. Furber

doi:10.1145/2016604.2016622

Maintaining real-time synchrony on SpiNNaker

Sergio Davies, Alexander D. Rast, Francesco Galluppi, Steve B. Furber

Advanced Processor Technology

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

Abstract

As an asynchronous universal multiprocessor for real-time neural simulation, SpiNNaker presents timing concerns not present in synchronous systems. In this paper we present a series of tools that solve the problem of synchronising a multichip distributed simulation containing multiple independent time domains. These tools hint at an important neural modelling capability of the SpiNNaker system: the ability to decouple the system time from the model time, leading to an abstract-time neural modelling platform. © 2011 Authors.

Original language	English
Title of host publication	Proceedings of the 8th ACM International Conference on Computing Frontiers, CF'11\|Proc. ACM Int. Conf. Comput. Front., CF
Place of Publication	New York, USA.
Publisher	Association for Computing Machinery
ISBN (Print)	9781450306980
DOIs	https://doi.org/10.1145/2016604.2016622
Publication status	Published - 2011
Event	8th ACM International Conference on Computing Frontiers, CF'11 - Ischia Duration: 1 Jul 2011 → …

Conference

Conference	8th ACM International Conference on Computing Frontiers, CF'11
City	Ischia
Period	1/07/11 → …

Keywords

asynchronous
event-driven
neural
SpiNNaker

Access to Document

10.1145/2016604.2016622

Cite this

@inproceedings{4235ff0034bb4343a208157709e0ddf1,

title = "Maintaining real-time synchrony on SpiNNaker",

abstract = "As an asynchronous universal multiprocessor for real-time neural simulation, SpiNNaker presents timing concerns not present in synchronous systems. In this paper we present a series of tools that solve the problem of synchronising a multichip distributed simulation containing multiple independent time domains. These tools hint at an important neural modelling capability of the SpiNNaker system: the ability to decouple the system time from the model time, leading to an abstract-time neural modelling platform. {\textcopyright} 2011 Authors.",

keywords = "asynchronous, event-driven, neural, SpiNNaker",

author = "Sergio Davies and Rast, {Alexander D.} and Francesco Galluppi and Furber, {Steve B.}",

year = "2011",

doi = "10.1145/2016604.2016622",

language = "English",

isbn = "9781450306980",

booktitle = "Proceedings of the 8th ACM International Conference on Computing Frontiers, CF'11|Proc. ACM Int. Conf. Comput. Front., CF",

publisher = "Association for Computing Machinery",

address = "United States",

note = "8th ACM International Conference on Computing Frontiers, CF'11 ; Conference date: 01-07-2011",

}

Davies, S, Rast, AD, Galluppi, F & Furber, SB 2011, Maintaining real-time synchrony on SpiNNaker. in Proceedings of the 8th ACM International Conference on Computing Frontiers, CF'11|Proc. ACM Int. Conf. Comput. Front., CF. Association for Computing Machinery, New York, USA., 8th ACM International Conference on Computing Frontiers, CF'11, Ischia, 1/07/11. https://doi.org/10.1145/2016604.2016622

Maintaining real-time synchrony on SpiNNaker. / Davies, Sergio; Rast, Alexander D.; Galluppi, Francesco et al.
Proceedings of the 8th ACM International Conference on Computing Frontiers, CF'11|Proc. ACM Int. Conf. Comput. Front., CF. New York, USA.: Association for Computing Machinery, 2011.

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

TY - GEN

T1 - Maintaining real-time synchrony on SpiNNaker

AU - Davies, Sergio

AU - Rast, Alexander D.

AU - Galluppi, Francesco

AU - Furber, Steve B.

PY - 2011

Y1 - 2011

N2 - As an asynchronous universal multiprocessor for real-time neural simulation, SpiNNaker presents timing concerns not present in synchronous systems. In this paper we present a series of tools that solve the problem of synchronising a multichip distributed simulation containing multiple independent time domains. These tools hint at an important neural modelling capability of the SpiNNaker system: the ability to decouple the system time from the model time, leading to an abstract-time neural modelling platform. © 2011 Authors.

AB - As an asynchronous universal multiprocessor for real-time neural simulation, SpiNNaker presents timing concerns not present in synchronous systems. In this paper we present a series of tools that solve the problem of synchronising a multichip distributed simulation containing multiple independent time domains. These tools hint at an important neural modelling capability of the SpiNNaker system: the ability to decouple the system time from the model time, leading to an abstract-time neural modelling platform. © 2011 Authors.

KW - asynchronous

KW - event-driven

KW - neural

KW - SpiNNaker

U2 - 10.1145/2016604.2016622

DO - 10.1145/2016604.2016622

M3 - Conference contribution

SN - 9781450306980

BT - Proceedings of the 8th ACM International Conference on Computing Frontiers, CF'11|Proc. ACM Int. Conf. Comput. Front., CF

PB - Association for Computing Machinery

CY - New York, USA.

T2 - 8th ACM International Conference on Computing Frontiers, CF'11

Y2 - 1 July 2011

ER -

Maintaining real-time synchrony on SpiNNaker

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this