Title :
CNN model on cell multiprocessor array
Author :
Nagy, Zoltín ; Kék, László ; Kincses, Zoltán ; Szolgay, Péter
Author_Institution :
Comput. & Autom. Res. Inst., Hungarian Acad. of Sci., Budapest
Abstract :
Array computers can be useful in the solution of numerical spatio-temporal problems such as the state equation of the CNN or partial differential equations (PDE). IBM has recently introduced the cell broadband engine (Cell BE) Architecture which contains 8 identical vector processors in an array structure. In the paper the implementation of CNN simulation kernel on the Cell BE is described. The simulation kernel is optimized, according to the special requirements of the Cell BE and can use linear and also nonlinear (piecewise linear) templates. The area/speed/power tradeoffs of our solution and different hardware implementations are also compared.
Keywords :
cellular neural nets; multiprocessing systems; parallel processing; partial differential equations; piecewise linear techniques; vector processor systems; CNN model; CNN simulation kernel; IBM; array computers; cell broadband engine architecture; cell multiprocessor array; partial differential equations; piecewise linear templates; spatio-temporal problems; state equation; vector processors; Blades; Cellular neural networks; Computer architecture; Concurrent computing; Distributed computing; Engines; Hardware; Kernel; Partial differential equations; Signal processing;
Conference_Titel :
Circuit Theory and Design, 2007. ECCTD 2007. 18th European Conference on
Conference_Location :
Seville
Print_ISBN :
978-1-4244-1341-6
Electronic_ISBN :
978-1-4244-1342-3
DOI :
10.1109/ECCTD.2007.4529590