A GALS Infrastructure for a Massively Parallel Multiprocessor
|
The Spinnaker (Spiking Neural Network Architecture) system for large-scale neural modeling is based on a scalable processor chip containing multiple ARM cores. Using a globally asynchronous, locally synchronous (GALS) approach allows custom, off-the-shelf IP to be readily integrated without significant timing-closure design effort. The ARM processors are used to simulate neurons, and generated neural events are carried over an on-chip, packet-switched fabric. This self-timed interconnect is also extended off chip to a provide chip-to-chip interconnect that scales to networks of thousands of chips.
[1] 454 S. Furber and S. Temple, "Neural Systems Engineering," J. Royal Society Interface, vol. 4, no. 13, Apr. 2007, pp. 193–206.
[2] J. Bainbridge and S. Furber, "Chain: A Delay-Insensitive Chip Area Interconnect," IEEE Micro, vol. 22, no. 5, Sept.-Oct. 2002, pp. 16–23.
[3] T. Verhoeff, "Delay-Insensitive Codes—An Overview," Distributed Computing, vol. 3, no. 1, Mar. 1988, pp. 1–8.
[4] Silistix Self-Timed Interconnect Technology, Silistix; http://www.silistix.comtechnology_silistix.php .
[5] Advanced Microcontroller Bus Architecture (AMBA) Specification, Rev. 2.0, ARM, May 1999, http://www.arm.com/products/solutionsAMBAHomePage.html .
[6] W.J. Bainbridge et al., "Delay-Insensitive, Point-to-Point Interconnect Using M-of-N Codes," Proc. 9th IEEE Int'l Symp. Asynchronous Circuits and Systems (ASYNC 03), IEEE CS Press, 2003, pp. 132–140.
[7] J. Wu and S. Furber, "Delay Insensitive Chip-to-Chip Interconnect Using Incomplete 2-of-7 NRZ Data Encoding," Proc. 18th UK Asynchronous Forum, University of Newcastle upon Tyne, 2006, pp. 16–19, http://async.org.ukukasyncforum18/.
[8] Y. Shi and S. Furber, "Error Checking and Resetting Mechanisms for Asynchronous Interconnect," Proc. 18th UK Asynchronous Forum, University of Newcastle upon Tyne, 2006, pp. 24–27, http://async.org.uk/ukasyncforum18.
[9] ARM968E-S, ARM, http://www.arm.com/products/CPUsARM968E-S.html .
Index Terms:
massively parallel multiprocessor, GALS, Spinnaker, neural modeling, self-timed interconnect
Citation:
Luis A. Plana, Steve B. Furber, Steve Temple, Mukaram Khan, Yebin Shi, Jian Wu, Shufan Yang, "A GALS Infrastructure for a Massively Parallel Multiprocessor," IEEE Design and Test of Computers, vol. 24, no. 5, pp. 454-463, Sept. 2007, doi:10.1109/MDT.2007.149
PURCHASE ARTICLE: $19
IEEE Xplore Subscribers
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb