Distributed configuration of massively-parallel simulation on SpiNNaker neuromorphic hardware

SpiNNaker is a massively-parallel neuromorphic computing architecture designed to model very large, biologically plausible spiking neural networks in real-time. A SpiNNaker machine consists of up to 216 homogeneous eighteen-core multiprocessor chips, each with an on-board router which forms links with neighbouring chips for packet-switched interprocessor communications. The architecture is designed for dynamic reconfiguration and optimised for transmission of neural activity data, which presents a challenge for machine configuration, program loading and simulation monitoring given a lack of globally-shared memory resources, intrinsic addressing mode or sideband configuration channel. We propose distributed software mechanisms to address these problems and present experiments which demonstrate the necessity of this approach in contrast to centralised mechanisms.

[1]  H. Markram.  The Blue Brain Project , 2006, Nature Reviews Neuroscience.

[2]  Steve B. Furber,et al.  Event-Driven Configuration of a Neural Network CMP System over a Homogeneous Interconnect Fabric , 2009, 2009 Eighth International Symposium on Parallel and Distributed Computing.

[3]  Steve B. Furber,et al.  A General-Purpose Model Translation System for a Universal Neural Chip , 2010, ICONIP.

[4]  Cameron Patterson,et al.  Event-driven configuration of a neural network CMP system over an homogeneous interconnect fabric , 2011, Parallel Comput..

[5]  Mark James,et al.  Design of Low-Cost, Real-Time Simulation Systems for Large Neural Networks , 1992, J. Parallel Distributed Comput..

[6]  Eugene M. Izhikevich,et al.  Dynamical Systems in Neuroscience: The Geometry of Excitability and Bursting , 2006 .

[7]  Ben Goertzel,et al.  A world survey of artificial brain projects, Part I: Large-scale brain simulations , 2010, Neurocomputing.

[8]  G. Edelman,et al.  Large-scale model of mammalian thalamocortical systems , 2008, Proceedings of the National Academy of Sciences.

[9]  Eugene M. Izhikevich,et al.  Simple model of spiking neurons , 2003, IEEE Trans. Neural Networks.

[10]  Thomas L. Sterling,et al.  BEOWULF: A Parallel Workstation for Scientific Computation , 1995, ICPP.

[11]  A. Thomson,et al.  Functional Maps of Neocortical Local Circuitry , 2007, Front. Neurosci..

[12]  Pierre Yger,et al.  PyNN: A Common Interface for Neuronal Network Simulators , 2008, Front. Neuroinform..

[13]  Dirk Hoenicke,et al.  Blue Gene/L compute chip: Control, test, and bring-up infrastructure , 2005, IBM J. Res. Dev..

[14]  R. Douglas,et al.  A Quantitative Map of the Circuit of Cat Primary Visual Cortex , 2004, The Journal of Neuroscience.

[15]  George L.-T. Chiu,et al.  Overview of the Blue Gene/L system architecture , 2005, IBM J. Res. Dev..

[16]  de GarisHugo,et al.  A world survey of artificial brain projects, Part II , 2010 .

[17]  Steve B. Furber,et al.  Understanding the interconnection network of SpiNNaker , 2009, ICS.

[18]  Stephen B. Furber,et al.  Efficient modelling of spiking neural networks on a scalable chip multiprocessor , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[19]  Örjan Ekeberg,et al.  Large-Scale Modeling – a Tool for Conquering the Complexity of the Brain , 2008, Frontiers Neuroinformatics.