Accelerating large-scale simulations of cortical neuronal network development
Computing & Software Systems Program, University of Washington Bothell, 18115 Campus Way NE, Bothell, WA 98011 USA
CSS Technical Report UWB-CSS-12-01, 2012
@article{kawasaki2012accelerating,
title={Accelerating large-scale simulations of cortical neuronal network development},
author={Kawasaki, F. and Stiber, M.},
year={2012}
}
Cultured dissociated cortical cells grown into networks on multi-electrode arrays are used to investigate neuronal network development, activity, plasticity, response to stimuli, the effects of pharmacological agents, etc. We made a computational model of such a neuronal network and studied the interplay of individual neuron activity, cell culture development, and network behavior. For small networks (100 neurons in a 10 x 10 arrangement), we concluded that our simulations’ behaviors were dominated by their limited size. However, increasing network size required huge computational resources: for a single-threaded simulator, a 100 x 100 neuron simulation would take at least 2,000 hours (83 days). To tackle this problem, we ported the network simulator to the GPU. A first, naive implementation performed about 2.4 times faster than the single thread simulator. By progressively modifying the simulator structure, we achieved about 23 times performance gain compared with the single threaded simulator, bringing large-scale simulations into the realm of feasibility. Through this experience, we identified factors that limit performance improvement for this type of application.
March 23, 2012 by hgpu
