No abstract

The development of physiologically plausible neuron models comes with increased complexity, which poses a challenge for many-core computing. In this work, we have chosen an extension of the demanding Hodgkin-Huxley model for the neurons of the Inferior Olivary Nucleus, an area of vital importance for motor skills. The computing fabric of choice is an Intel Xeon-Xeon Phi system, widely-used in modern computing infrastructure. The target application is parallelized with combinations of MPI and OpenMP. The best configurations are scaled up to human InfOli numbers.

No abstract

Detailed brain modeling has been presenting significant challenges to the world of high-performance computing (HPC), posing computational problems that can benefit from modern hardware-acceleration technologies. We explore the capacity of GPUs for simulating large-scale neuronal networks based on the Adaptive Exponential neuron-model, which is widely used in the neuroscientific community. Our GPU-powered simulator acts as a benchmark to evaluate the strengths and limitations of modern GPUs, as well as to explore their scaling properties when simulating large neural networks. This work presents an optimized GPU implementation that outperforms a reference multicore implementation by 50×, whereas utilizing a dual-GPU configuration can deliver a speedup of 90× for networks of 20, 000 fully interconnected AdEx neurons.

Research on the prevention of epileptic seizures has led to approaches for future treatment techniques, which rely on the demanding computation of generalized partial directed coherence (GPDC) on electroencephalogram (EEG) data. A fast computation of such metrics is a key factor both for the off-line optimization of algorithmic parameters and for its real-time implementation. Aiming at speeding up the GPDC computations on EEG data, the current paper presents massively parallel computational strategies for implementing the GPDC on many-core architectures. We apply the proposed strategies on commercial and experimental many-core platforms and we compare the results of the computation time of a set of EEG data on the Bulldozer and Ivy Bridge x86_64 serial processors. We test the GPUs of nVidia GTX 550 Ti and GTX 670, which at the best case achieve a significant speedup of 190x and 460x respectively. Moreover, we apply the proposed parallelization strategies on the Single-Chip Cloud Computer (SCC), an experimental processor created by Intel Labs