Overview of the SpiNNaker System Architecture

SpiNNaker (a contraction of Spiking Neural Network Architecture) is a million-core computing engine whose flagship goal is to be able to simulate the behavior of aggregates of up to a billion neurons in real time. It consists of an array of ARM9 cores, communicating via packets carried by a custom interconnect fabric. The packets are small (40 or 72 bits), and their transmission is brokered entirely by hardware, giving the overall engine an extremely high bisection bandwidth of over 5 billion packets/s. Three of the principal axioms of parallel machine design (memory coherence, synchronicity, and determinism) have been discarded in the design without, surprisingly, compromising the ability to perform meaningful computations. A further attribute of the system is the acknowledgment, from the initial design stages, that the sheer size of the implementation will make component failures an inevitable aspect of day-to-day operation, and fault detection and recovery mechanisms have been built into the system at many levels of abstraction. This paper describes the architecture of the machine and outlines the underlying design philosophy; software and applications are to be described in detail elsewhere, and only introduced in passing here as necessary to illuminate the description.

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

[2]  Steve B. Furber,et al.  Modeling Spiking Neural Networks on SpiNNaker , 2010, Computing in Science & Engineering.

[3]  B. Pakkenberg,et al.  Aging and the human neocortex , 2003, Experimental Gerontology.

[4]  Steve B. Furber,et al.  Event-Driven Simulation of Arbitrary Spiking Neural Networks on SpiNNaker , 2011, ICONIP.

[5]  廣瀬雄一,et al.  Neuroscience , 2019, Workplace Attachments.

[6]  Misha A. Mahowald,et al.  An Analog VLSI System for Stereoscopic Vision , 1994 .

[7]  Jim D. Garside,et al.  SpiNNaker: Design and Implementation of a GALS Multicore System-on-Chip , 2011, JETC.

[8]  Lizandro D. Solano-Quinde,et al.  Module Prototype for Online Failure Prediction for the IBM Blue Gene/L , 2008, 2008 IEEE International Conference on Electro/Information Technology.

[9]  C. U. M. Smith,et al.  Elements of molecular neurobiology , 1989 .

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

[11]  Steve B. Furber,et al.  Neural Systems Engineering , 2008, Computational Intelligence: A Compendium.

[12]  Israel Koren,et al.  Fault-Tolerant Systems , 2007 .

[13]  Luis A. Plana,et al.  A GALS Infrastructure for a Massively Parallel Multiprocessor , 2007, IEEE Design & Test of Computers.

[14]  Steve B. Furber,et al.  A Multicast Routing Scheme for a Universal Spiking Neural Network Architecture , 2010, Comput. J..

[15]  Ibm Blue,et al.  Overview of the IBM Blue Gene/P Project , 2008, IBM J. Res. Dev..

[16]  L. Gould.  The Cat is Out of the Bag , 1996 .

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

[18]  Massimo A. Sivilotti,et al.  Wiring considerations in analog VLSI systems, with application to field-programmable networks , 1992 .

[19]  John Bainbridge.  The CHAIN Works Tool Suite: A Complete Industrial Design Flow for Networks-on-Chips , 2009 .

[20]  Dharmendra S. Modha,et al.  The cat is out of the bag: cortical simulations with 109 neurons, 1013 synapses , 2009, Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis.