Interconnection system for the SpiNNaker biologically inspired multi-computer
Interconnection system for the SpiNNaker biologically inspired multi-computer
SpiNNaker is a large-scale multi-core computing engine designed to model heavily distributed fine-grain problems. The machine is constructed hierarchically: 1 monitor and 16 worker processors form a single node of a toroidal compute 'surface'. The six high-speed bi-directional links of each node are used for triangular edge connections that provide alternative routes around problematic links. The system itself is scalable from one node up to 216 resulting in a maximum of 220 worker processors. SpiNNaker is an isotropic homogeneous network of processors that deliberately includes no central overseer. A consequence of this isotropy is an absence of perimeter and hence no natural position for peripheral I/O connections. This study describes the practical techniques and details employed in two components of the system: (a) SpiNNlink is the proposed board-to-board interconnection system that multiplexes 48 separate 250 Mbps SpiNNaker links through six off-board connections without compromising the overall system bisection bandwidth, forms an isotropic metanetwork on top of SpiNNaker without requiring any cooperation from system software, and remains transparent to the SpiNNaker network; and (b) SpiNNterceptor is the proposed peripheral I/O subsystem developed as a layer on top of SpiNNlink that provides over 18 Gbps of minimally disruptive communication between SpiNNaker applications and externally connected equipment.
Dugan, K.
90895e3c-ecf4-46d7-b453-395eec184b1e
Reeve, J.S.
dd909010-7d44-44ea-83fe-a09e4d492618
Brown, A.
5c19e523-65ec-499b-9e7c-91522017d7e0
Furber, S.
138aa65c-75e3-4c42-bd9c-2d17d9b61ded
24 June 2013
Dugan, K.
90895e3c-ecf4-46d7-b453-395eec184b1e
Reeve, J.S.
dd909010-7d44-44ea-83fe-a09e4d492618
Brown, A.
5c19e523-65ec-499b-9e7c-91522017d7e0
Furber, S.
138aa65c-75e3-4c42-bd9c-2d17d9b61ded
Dugan, K., Reeve, J.S., Brown, A. and Furber, S.
(2013)
Interconnection system for the SpiNNaker biologically inspired multi-computer.
IET Computers & Digital Techniques, 7 (3).
(doi:10.1049/iet-cdt.2012.0139).
Abstract
SpiNNaker is a large-scale multi-core computing engine designed to model heavily distributed fine-grain problems. The machine is constructed hierarchically: 1 monitor and 16 worker processors form a single node of a toroidal compute 'surface'. The six high-speed bi-directional links of each node are used for triangular edge connections that provide alternative routes around problematic links. The system itself is scalable from one node up to 216 resulting in a maximum of 220 worker processors. SpiNNaker is an isotropic homogeneous network of processors that deliberately includes no central overseer. A consequence of this isotropy is an absence of perimeter and hence no natural position for peripheral I/O connections. This study describes the practical techniques and details employed in two components of the system: (a) SpiNNlink is the proposed board-to-board interconnection system that multiplexes 48 separate 250 Mbps SpiNNaker links through six off-board connections without compromising the overall system bisection bandwidth, forms an isotropic metanetwork on top of SpiNNaker without requiring any cooperation from system software, and remains transparent to the SpiNNaker network; and (b) SpiNNterceptor is the proposed peripheral I/O subsystem developed as a layer on top of SpiNNlink that provides over 18 Gbps of minimally disruptive communication between SpiNNaker applications and externally connected equipment.
Text
cdt-2012-final-02.pdf
- Author's Original
Restricted to Registered users only
Request a copy
More information
e-pub ahead of print date: May 2013
Published date: 24 June 2013
Organisations:
EEE
Identifiers
Local EPrints ID: 350496
URI: http://eprints.soton.ac.uk/id/eprint/350496
PURE UUID: 4abbc433-1788-4786-ada7-611c57de2c05
Catalogue record
Date deposited: 04 Apr 2013 14:44
Last modified: 14 Mar 2024 13:26
Export record
Altmetrics
Contributors
Author:
K. Dugan
Author:
J.S. Reeve
Author:
A. Brown
Author:
S. Furber
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics