• DocumentCode
    1955418
  • Title

    Biologically-Inspired Massively-Parallel Architectures - Computing Beyond a Million Processors

  • Author

    Furber, Steve ; Brown, Andrew

  • Author_Institution
    Univ. of Manchester, Manchester, UK
  • fYear
    2009
  • fDate
    1-3 July 2009
  • Firstpage
    3
  • Lastpage
    12
  • Abstract
    The SpiNNaker project aims to develop parallel computer systems with more than a million embedded processors. The goal of the project is to support large-scale simulations of systems of spiking neurons in biological real time, an application that is highly parallel but also places very high loads on the communication infrastructure due to the very high connectivity of biological neurons. The scale of the machine requires fault-tolerance and power-efficiency to influence the design throughout, and the development has resulted in innovation at every level of design, including a self-timed inter-chip communication system that is resistant to glitch-induced deadlock and `emergency´ hardware packet re-routing around failed inter-chip links, through to run-time support for functional migration and real-time fault mitigation.
  • Keywords
    biology computing; microprocessor chips; neural nets; parallel architectures; SpiNNaker project; biological neurons; biologically-inspired massively-parallel architectures; emergency hardware packet re-routing; fault tolerance; functional migration; glitch-induced deadlock; million embedded processors; parallel computer systems; power efficiency; real-time fault mitigation; self-timed inter-chip communication system; spiking neurons; Application software; Biological system modeling; Biology computing; Computational modeling; Computer architecture; Concurrent computing; Embedded computing; Large-scale systems; Neurons; Real time systems; Massively-parallel; fault-tolerance; neural networks;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Application of Concurrency to System Design, 2009. ACSD '09. Ninth International Conference on
  • Conference_Location
    Augsburg
  • ISSN
    1550-4808
  • Print_ISBN
    978-0-7695-3697-2
  • Type

    conf

  • DOI
    10.1109/ACSD.2009.17
  • Filename
    5291092