1. Gaussian and Exponential Lateral Connectivity on Distributed Spiking Neural Network Simulation
- Author
-
Paolo Cretaro, Elena Pastorelli, Roberto Ammendola, Giulia De Bonis, Pierluigi Paolucci, Francesco Simula, Andrea Biagioni, Fabrizio Capuani, Piero Vicini, Alessandro Lonardo, Francesca Lo Cicero, Luca Pontisso, and Michele Martinelli
- Subjects
FOS: Computer and information sciences ,0301 basic medicine ,Computer science ,Gaussian ,InfiniBand ,Parallel computing ,lateral synaptic connectivity ,spiking neural network ,Cortical Simulation ,distributed computing ,03 medical and health sciences ,symbols.namesake ,0302 clinical medicine ,Exponential growth ,Spike frequency ,hard- ware/software co-design ,Neural and Evolutionary Computing (cs.NE) ,Scaling ,Spiking neural network ,Xeon ,Computer Science - Neural and Evolutionary Computing ,Exponential function ,030104 developmental biology ,Computer Science - Distributed, Parallel, and Cluster Computing ,Cortical simulation ,Distributed Computing ,Hardware/software co-design ,Lateral synaptic connectivity ,Spiking Neural Network ,Computer Networks and Communications ,Hardware and Architecture ,FOS: Biological sciences ,Quantitative Biology - Neurons and Cognition ,symbols ,Neurons and Cognition (q-bio.NC) ,Distributed, Parallel, and Cluster Computing (cs.DC) ,030217 neurology & neurosurgery - Abstract
We measured the impact of long-range exponentially decaying intra-areal lateral connectivity on the scaling and memory occupation of a distributed spiking neural network simulator compared to that of short-range Gaussian decays. While previous studies adopted short-range connectivity, recent experimental neurosciences studies are pointing out the role of longer-range intra-areal connectivity with implications on neural simulation platforms. Two-dimensional grids of cortical columns composed by up to 11 M point-like spiking neurons with spike frequency adaption were connected by up to 30 G synapses using short- and long-range connectivity models. The MPI processes composing the distributed simulator were run on up to 1024 hardware cores, hosted on a 64 nodes server platform. The hardware platform was a cluster of IBM NX360 M5 16-core compute nodes, each one containing two Intel Xeon Haswell 8-core E5-2630 v3 processors, with a clock of 2.40 G Hz, interconnected through an InfiniBand network, equipped with 4x QDR switches., 9 pages, 9 figures, added reference to final peer reviewed version on conference paper and DOI
- Published
- 2018
- Full Text
- View/download PDF