Biologically-Inspired Massively Parallel Architectures - computing beyond a million processors
Lead Research Organisation:
University of Manchester
Department Name: Computer Science
Abstract
The human brain remains as one of the great frontiers of science - how does this organ upon which we all depend so critically actually do its job? A great deal is known about the underlying technology - the neuron - and we can observe large-scale brain activity through techniques such as magnetic resonance imaging, but this knowledge barely starts to tell us how the brain works. Something is happening at the intermediate levels of processing that we have yet to begin to understand, but the essence of the brain's information processing function probably lies in these intermediate levels. To get at these middle layers requires that we build models of very large systems of spiking neurons, with structures inspired by the increasingly detailed findings of neuroscience, in order to investigate the emergent behaviours, adaptability and fault-tolerance of those systems.Our goal in this project is to deliver machines of unprecedented cost-effectiveness for this task, and to make them readily accessible to as wide a user base as possible. We will also explore the applicability of the unique architecture that has emerged from the pursuit of this goal to other important application domains.
Publications
Liu C
(2018)
Memory-Efficient Deep Learning on a SpiNNaker 2 Prototype.
in Frontiers in neuroscience
Liu Q
(2016)
Benchmarking Spike-Based Visual Recognition: A Dataset and Evaluation.
in Frontiers in neuroscience
Mikaitis M
(2018)
Neuromodulated Synaptic Plasticity on the SpiNNaker Neuromorphic System
in Frontiers in Neuroscience
Navaridas J
(2015)
SpiNNaker: Enhanced multicast routing
in Parallel Computing
Navaridas J
(2013)
SpiNNaker: Fault tolerance in a power- and area- constrained large-scale neuromimetic architecture
in Parallel Computing
Navaridas J
(2014)
On generating multicast routes for SpiNNaker
Painkras E
(2013)
SpiNNaker: A 1-W 18-Core System-on-Chip for Massively-Parallel Neural Network Simulation
in IEEE Journal of Solid-State Circuits
Patterson C
(2012)
Visualising large-scale neural network models in real-time
Patterson C
(2012)
Scalable communications for a million-core neural processing architecture
in Journal of Parallel and Distributed Computing
Patterson C
(2012)
Managing a Massively-Parallel Resource-Constrained Computing Architecture
Peres L
(2022)
Parallelization of Neural Processing on Neuromorphic Hardware
in Frontiers in Neuroscience
Plana L
(2020)
spiNNlink: FPGA-Based Interconnect for the Million-Core SpiNNaker System
in IEEE Access
Plana L
(2011)
SpiNNaker Design and Implementation of a GALS Multicore System-on-Chip
in ACM Journal on Emerging Technologies in Computing Systems
Rast A
(2011)
Concurrent heterogeneous neural model simulation on real-time neuromimetic hardware.
in Neural networks : the official journal of the International Neural Network Society
Rast A
(2010)
Scalable event-driven native parallel processing
Rast A
(2011)
Managing Burstiness and Scalability in Event-Driven Models on the SpiNNaker Neuromimetic System
in International Journal of Parallel Programming
Rast A
(2015)
Neural Information Processing
Rast AD
(2018)
Behavioral Learning in a Cognitive Neuromorphic Robot: An Integrative Approach.
in IEEE transactions on neural networks and learning systems
Rhodes O
(2020)
Real-time cortical simulation on neuromorphic hardware.
in Philosophical transactions. Series A, Mathematical, physical, and engineering sciences
Rhodes O
(2018)
sPyNNaker: A Software Package for Running PyNN Simulations on SpiNNaker.
in Frontiers in neuroscience
Rowley AGD
(2019)
SpiNNTools: The Execution Engine for the SpiNNaker Platform.
in Frontiers in neuroscience
Sen-Bhattacharya B
(2017)
A Spiking Neural Network Model of the Lateral Geniculate Nucleus on the SpiNNaker Machine.
in Frontiers in neuroscience
Sen-Bhattacharya B
(2018)
Building a Spiking Neural Network Model of the Basal Ganglia on SpiNNaker
in IEEE Transactions on Cognitive and Developmental Systems
Serrano-Gotarredona T
(2015)
ConvNets experiments on SpiNNaker
Sharp T
(2013)
Correctness and performance of the SpiNNaker architecture
Sharp T
(2014)
Real-time million-synapse simulation of rat barrel cortex.
in Frontiers in neuroscience
Sharp T
(2012)
Power-efficient simulation of detailed cortical microcircuits on SpiNNaker.
in Journal of neuroscience methods
Stromatias E
(2013)
Power analysis of large-scale, real-time neural networks on SpiNNaker
Stromatias E
(2015)
Robustness of spiking Deep Belief Networks to noise and reduced bit precision of neuro-inspired hardware platforms.
in Frontiers in neuroscience
Stromatias E
(2014)
Optimising the overall power usage on the SpiNNaker neuromimetic platform
Sugiarto I
(2017)
Optimized task graph mapping on a many-core neuromorphic supercomputer
Sugiarto I
(2021)
Fine-grained or coarse-grained? Strategies for implementing parallel genetic algorithms in a programmable neuromorphic platform
in TELKOMNIKA (Telecommunication Computing Electronics and Control)
Sugiarto, I.
(2017)
Task Graph Mapping of General Purpose Applications on a Neuromorphic Platform
Title | Talk |
Description | 'Talk' is an exhibit by Swedish artist Tove Kjellmark built as part of "The Imitation Game" exhibition at Manchester Art Gallery for part of Manchester's role as 2016 European City of Science. 'Talk' is a robotic exhibit that uses SpiNNaker technology. |
Type Of Art | Artwork |
Year Produced | 2016 |
Impact | The exhibition runs from February to June 2016. |
URL | http://manchesterartgallery.org/exhibitions-and-events/exhibition/the-imitation-game/ |
Description | A specially-designed passively-parallel computer can support real-time brain models better than a high-end supercomputer. |
Exploitation Route | The SpiNNaker platform can be used to explore models of brain function, neurorobotics, etc. |
Sectors | Digital/Communication/Information Technologies (including Software),Education,Electronics,Healthcare |
URL | http://apt.cs.manchester.ac.uk/projects/SpiNNaker/ |
Description | We have built the SpiNNaker real-time brain-modelling machine, to be used in the EU ICT Flagship Human Brain Project. Some 90 small SpiNNaker boards are out on loan to labs around the world. The University has also made sales of SpiNNaker machines to academic, government and industry research labs world-wide. |
First Year Of Impact | 2015 |
Sector | Digital/Communication/Information Technologies (including Software),Electronics |
Description | (BIMPC) - Biologically-Inspired Massively-Parallel Computation |
Amount | € 2,399,761 (EUR) |
Funding ID | 320689 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 03/2013 |
End | 02/2018 |
Description | (HBP SGA1) - Human Brain Project Specific Grant Agreement 1 |
Amount | € 89,000,000 (EUR) |
Funding ID | 720270 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 04/2016 |
End | 03/2018 |
Description | (HBP SGA2) - Human Brain Project Specific Grant Agreement 2 |
Amount | € 88,000,000 (EUR) |
Funding ID | 785907 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 04/2018 |
End | 03/2020 |
Description | (HBP SGA3) - Human Brain Project Specific Grant Agreement 3 |
Amount | € 150,000,000 (EUR) |
Funding ID | 945539 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 04/2020 |
End | 03/2023 |
Description | (HBP) - Human Brain Project |
Amount | € 1,632,174 (EUR) |
Funding ID | 284941 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 05/2011 |
End | 04/2012 |
Description | (HBP) - The Human Brain Project |
Amount | € 72,522,840 (EUR) |
Funding ID | 604102 |
Organisation | European Commission |
Sector | Public |
Country | European Union (EU) |
Start | 10/2013 |
End | 02/2017 |
Description | A R M Ltd |
Organisation | Arm Limited |
Country | United Kingdom |
Sector | Private |
Start Year | 2006 |
Description | Silistix Ltd |
Organisation | Silistix Ltd |
Country | United Kingdom |
Sector | Private |
Start Year | 2006 |
Title | Multicast Communications Router |
Description | This is the key patent for the multicast packet-switch spike packet routing used on SpiNNaker. |
IP Reference | GB0524126.0 |
Protection | Patent granted |
Year Protection Granted | |
Licensed | Yes |
Impact | SpiNNaker has been loaned and sold to about 75 research groups around the world. |
Company Name | Cogniscience Ltd |
Description | Owns the SpiNNaker IP |
Impact | Receives royalties on SpiNNaker sales. |