Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation

Jaco Hofman; Amir Zjajo; Carlo Galuzzi; Rene van Leuken

doi:10.1109/embc.2016.7592053

Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation

Jaco Hofman, Amir Zjajo, Carlo Galuzzi, Rene van Leuken

Signal Processing Systems

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

5 Downloads (Pure)

Abstract

State-of-the-art neuron simulators are capable of simulating at most few tens/hundreds of neurons in real-time due to the exponential growth in the communication costs with the number of simulated neurons. In this paper, we present a novel, reconfigurable, multi-chip system architecture based on localized communication, which effectively reduces the communication cost to a linear growth. The system is very flexible and it allows to tune, at run-time, various parameters, e.g. the intracellular concentration of chemical compounds, the interconnection scheme between the neurons. Experimental results indicate that the proposed system architecture allows the simulation of up to few thousands biophysically accurate neurons over multiple chips.

Original language	English
Title of host publication	Proceedings - 38th Annual IEEE International Conference of the Engineering in Medicine and Biology Society (EMBS)
Editors	J. Patton
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	5829-5832
Number of pages	4
ISBN (Electronic)	978-1-4577-0220-4
ISBN (Print)	978-1-4577-0219-8
DOIs	https://doi.org/10.1109/embc.2016.7592053
Publication status	Published - 2016
Event	38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016 - Orlando, Florida, United States Duration: 16 Aug 2016 → 20 Aug 2016 http://embc.embs.org/2016/

Conference

Conference	38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016
Abbreviated title	EMBC
Country/Territory	United States
City	Orlando, Florida
Period	16/08/16 → 20/08/16
Internet address	http://embc.embs.org/2016/

Keywords

Field programmable gate arrays
Neurons
Network topology
Integrated circuit interconnections
Topology
Routing
Biological system modeling

Access to Document

10.1109/embc.2016.7592053

Cite this

@inproceedings{b9edf43b42fa458d823929093df27f81,

title = "Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation",

abstract = "State-of-the-art neuron simulators are capable of simulating at most few tens/hundreds of neurons in real-time due to the exponential growth in the communication costs with the number of simulated neurons. In this paper, we present a novel, reconfigurable, multi-chip system architecture based on localized communication, which effectively reduces the communication cost to a linear growth. The system is very flexible and it allows to tune, at run-time, various parameters, e.g. the intracellular concentration of chemical compounds, the interconnection scheme between the neurons. Experimental results indicate that the proposed system architecture allows the simulation of up to few thousands biophysically accurate neurons over multiple chips.",

keywords = "Field programmable gate arrays, Neurons, Network topology, Integrated circuit interconnections, Topology, Routing, Biological system modeling",

author = "Jaco Hofman and Amir Zjajo and Carlo Galuzzi and {van Leuken}, Rene",

year = "2016",

doi = "10.1109/embc.2016.7592053",

language = "English",

isbn = "978-1-4577-0219-8",

pages = "5829--5832",

editor = "J. Patton",

booktitle = "Proceedings - 38th Annual IEEE International Conference of the Engineering in Medicine and Biology Society (EMBS)",

publisher = "IEEE",

address = "United States",

note = "38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016, EMBC ; Conference date: 16-08-2016 Through 20-08-2016",

url = "http://embc.embs.org/2016/",

}

Hofman, J, Zjajo, A, Galuzzi, C & van Leuken, R 2016, Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation. in J Patton (ed.), Proceedings - 38th Annual IEEE International Conference of the Engineering in Medicine and Biology Society (EMBS). IEEE, Piscataway, NJ, pp. 5829-5832, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016, Orlando, Florida, United States, 16/08/16. https://doi.org/10.1109/embc.2016.7592053

Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation. / Hofman, Jaco; Zjajo, Amir; Galuzzi, Carlo et al.
Proceedings - 38th Annual IEEE International Conference of the Engineering in Medicine and Biology Society (EMBS). ed. / J. Patton. Piscataway, NJ: IEEE, 2016. p. 5829-5832.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation

AU - Hofman, Jaco

AU - Zjajo, Amir

AU - Galuzzi, Carlo

AU - van Leuken, Rene

PY - 2016

Y1 - 2016

N2 - State-of-the-art neuron simulators are capable of simulating at most few tens/hundreds of neurons in real-time due to the exponential growth in the communication costs with the number of simulated neurons. In this paper, we present a novel, reconfigurable, multi-chip system architecture based on localized communication, which effectively reduces the communication cost to a linear growth. The system is very flexible and it allows to tune, at run-time, various parameters, e.g. the intracellular concentration of chemical compounds, the interconnection scheme between the neurons. Experimental results indicate that the proposed system architecture allows the simulation of up to few thousands biophysically accurate neurons over multiple chips.

AB - State-of-the-art neuron simulators are capable of simulating at most few tens/hundreds of neurons in real-time due to the exponential growth in the communication costs with the number of simulated neurons. In this paper, we present a novel, reconfigurable, multi-chip system architecture based on localized communication, which effectively reduces the communication cost to a linear growth. The system is very flexible and it allows to tune, at run-time, various parameters, e.g. the intracellular concentration of chemical compounds, the interconnection scheme between the neurons. Experimental results indicate that the proposed system architecture allows the simulation of up to few thousands biophysically accurate neurons over multiple chips.

KW - Field programmable gate arrays

KW - Neurons

KW - Network topology

KW - Integrated circuit interconnections

KW - Topology

KW - Routing

KW - Biological system modeling

U2 - 10.1109/embc.2016.7592053

DO - 10.1109/embc.2016.7592053

M3 - Conference contribution

SN - 978-1-4577-0219-8

SP - 5829

EP - 5832

BT - Proceedings - 38th Annual IEEE International Conference of the Engineering in Medicine and Biology Society (EMBS)

A2 - Patton, J.

PB - IEEE

CY - Piscataway, NJ

T2 - 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016

Y2 - 16 August 2016 through 20 August 2016

ER -

Multi-Chip Dataflow Architecture for Massive Scale Biophysically Accurate Neuron Simulation

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this