Accelerating complex brain-model simulations on GPU platforms

Author

Nguyen, H. A. Du ; Al-Ars, Zaid ; Smaragdos, Georgios ; Strydis, Christos

Author_Institution

Lab. of Comput. Eng. Fac. of EE, Delft Univ. of Technol., Delft, Netherlands

fYear

2015

fDate

9-13 March 2015

Firstpage

974

Lastpage

979

Abstract

The Inferior Olive (IO) in the brain, in conjunction with the cerebellum, is responsible for crucial sensorimotor-integration functions in humans. In this paper, we simulate a computationally challenging IO neuron model consisting of three compartments per neuron in a network arrangement on GPU platforms. Several GPU platforms of the two latest NVIDIA GPU architectures (Fermi, Kepler) have been used to simulate large-scale IO-neuron networks. These networks have been ported on 4 diverse GPU platforms and implementation has been optimized, scoring 3x speedups compared to its unoptimized version. The effect of GPU L1-cache and thread block size as well as the impact of numerical precision of the application on performance have been evaluated and best configurations have been chosen. In effect, a maximum speedup of 160x has been achieved with respect to a reference CPU platform.

Keywords

brain; graphics processing units; medical computing; parallel architectures; Fermi architecture; GPU L1-cache; GPU platforms; Kepler architecture; NVIDIA GPU architectures; cerebellum; complex brain-model simulations; inferior olive; large-scale IO-neuron network simulation; network arrangement; numerical precision; performance evaluation; reference CPU platform; sensorimotor-integration functions; thread block size; Brain modeling; Computational modeling; Computer architecture; Graphics processing units; Instruction sets; Mathematical model; Nerve fibers;

fLanguage

English

Publisher

ieee

Conference_Titel

Design, Automation & Test in Europe Conference & Exhibition (DATE), 2015

Conference_Location

Grenoble

Print_ISBN

978-3-9815-3704-8

Type

conf

Filename

7092530