Granular layEr Simulator: Design and Multi-GPU Simulation of the Cerebellar Granular Layer

In modern computational modeling, neuroscientists need to reproduce long-lasting activity of large-scale networks, where neurons are described by highly complex mathematical models. These aspects strongly increase the computational load of the simulations, which can be efficiently performed by exploiting parallel systems to reduce the processing times. Graphics Processing Unit (GPU) devices meet this need providing on desktop High Performance Computing. In this work, authors describe a novel Granular layEr Simulator development implemented on a multi-GPU system capable of reconstructing the cerebellar granular layer in a 3D space and reproducing its neuronal activity. The reconstruction is characterized by a high level of novelty and realism considering axonal/dendritic field geometries, oriented in the 3D space, and following convergence/divergence rates provided in literature. Neurons are modeled using Hodgkin and Huxley representations. The network is validated by reproducing typical behaviors which are well-documented in the literature, such as the center-surround organization. The reconstruction of a network, whose volume is 600 × 150 × 1,200 μm3 with 432,000 granules, 972 Golgi cells, 32,399 glomeruli, and 4,051 mossy fibers, takes 235 s on an Intel i9 processor. The 10 s activity reproduction takes only 4.34 and 3.37 h exploiting a single and multi-GPU desktop system (with one or two NVIDIA RTX 2080 GPU, respectively). Moreover, the code takes only 3.52 and 2.44 h if run on one or two NVIDIA V100 GPU, respectively. The relevant speedups reached (up to ~38× in the single-GPU version, and ~55× in the multi-GPU) clearly demonstrate that the GPU technology is highly suitable for realistic large network simulations.

[1]  Egidio D'Angelo,et al.  Computational Reconstruction of Pacemaking and Intrinsic Electroresponsiveness in Cerebellar Golgi Cells , 2007, Frontiers in cellular neuroscience.

[2]  Chaitanya Medini,et al.  Reconstruction and Simulation of a Scaffold Model of the Cerebellar Network , 2019 .

[3]  Mancia Anguita,et al.  A Spiking Neural Simulator Integrating Event-Driven and Time-Driven Computation Schemes Using Parallel CPU-GPU Co-Processing: A Case Study , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[4]  Egidio D'Angelo,et al.  Frontiers in Cellular Neuroscience Cellular Neuroscience , 2022 .

[5]  R. Silver,et al.  Rapid Desynchronization of an Electrically Coupled Interneuron Network with Sparse Excitatory Synaptic Input , 2010, Neuron.

[6]  Chaitanya Medini,et al.  Reconstruction and Simulation of a Scaffold Model of the Cerebellar Network , 2019, bioRxiv.

[7]  Nikil D. Dutt,et al.  Efficient Spiking Neural Network Model of Pattern Motion Selectivity in Visual Cortex , 2014, Neuroinformatics.

[8]  A. Hodgkin,et al.  A quantitative description of membrane current and its application to conduction and excitation in nerve , 1952, The Journal of physiology.

[9]  Georgios Smaragdos,et al.  Exploring Complex Brain-Simulation Workloads on Multi-GPU Deployments , 2019, ACM Trans. Archit. Code Optim..

[10]  Detlef Heck,et al.  Detection of sequences in the cerebellar cortex: numerical estimate of the possible number of tidal-wave inducing sequences represented , 2003, Journal of Physiology-Paris.

[11]  Egidio D'Angelo,et al.  The Spatial Organization of Long-Term Synaptic Plasticity at the Input Stage of Cerebellum , 2007, The Journal of Neuroscience.

[12]  Nikil D. Dutt,et al.  CARLsim 4: An Open Source Library for Large Scale, Biologically Detailed Spiking Neural Network Simulation using Heterogeneous Clusters , 2018, 2018 International Joint Conference on Neural Networks (IJCNN).

[13]  Kristofer E. Bouchard,et al.  High-Performance Computing in Neuroscience for Data-Driven Discovery, Integration, and Dissemination , 2016, Neuron.

[14]  Giovanni Danese,et al.  The Human Brain Project: Parallel technologies for biologically accurate simulation of Granule cells , 2016, Microprocess. Microsystems.

[15]  R Angus Silver,et al.  Synaptic and Cellular Properties of the Feedforward Inhibitory Circuit within the Input Layer of the Cerebellar Cortex , 2008, The Journal of Neuroscience.

[16]  Tadashi Yamazaki,et al.  Real-time simulation of a cat-scale artificial cerebellum on PEZY-SC processors , 2019, Int. J. High Perform. Comput. Appl..

[17]  Huxley Af,et al.  A quantitative description of membrane current and its application to conduction and excitation in nerve. 1952. , 1990 .

[18]  Francesco Cremonesi,et al.  Understanding Computational Costs of Cellular-Level Brain Tissue Simulations Through Analytical Performance Models , 2020, Neuroinformatics.

[19]  Giovanni Danese,et al.  High Performant Simulations of Cerebellar Golgi Cells Activity , 2017, 2017 25th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP).

[20]  V Taglietti,et al.  Theta-Frequency Bursting and Resonance in Cerebellar Granule Cells: Experimental Evidence and Modeling of a Slow K+-Dependent Mechanism , 2001, The Journal of Neuroscience.

[21]  D. Rossi,et al.  Spillover-Mediated Transmission at Inhibitory Synapses Promoted by High Affinity α6 Subunit GABAA Receptors and Glomerular Geometry , 1998, Neuron.

[22]  Francesco Leporati,et al.  GPU Parallelization of Realistic Purkinje Cells with Complex Morphology , 2019, 2019 27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP).

[23]  Arne Møller,et al.  Total numbers of various cell types in rat cerebellar cortex estimated using an unbiased stereological method , 1993, Brain Research.

[24]  E. D’Angelo,et al.  Regulation of output spike patterns by phasic inhibition in cerebellar granule cells , 2014, Front. Cell. Neurosci..

[25]  Murray Shanahan,et al.  Three Tools for the Real-Time Simulation of Embodied Spiking Neural Networks Using GPUs , 2013, Neuroinformatics.

[26]  Eduardo Ros,et al.  Modeling the Cerebellar Microcircuit: New Strategies for a Long-Standing Issue , 2016, Front. Cell. Neurosci..

[27]  Thomas Nowotny,et al.  GeNN: a code generation framework for accelerated brain simulations , 2016, Scientific Reports.

[28]  Thierry Nieus,et al.  LTP regulates burst initiation and frequency at mossy fiber-granule cell synapses of rat cerebellum: experimental observations and theoretical predictions. , 2006, Journal of neurophysiology.

[29]  Eugene M. Izhikevich,et al.  Which model to use for cortical spiking neurons? , 2004, IEEE Transactions on Neural Networks.

[30]  Maria Diamantaki,et al.  Exploring the significance of morphological diversity for cerebellar granule cell excitability , 2017, Scientific Reports.

[31]  Egidio D'Angelo,et al.  Integration and regulation of glomerular inhibition in the cerebellar granular layer circuit , 2014, Front. Cell. Neurosci..

[32]  Thierry Nieus,et al.  A Realistic Large-Scale Model of the Cerebellum Granular Layer Predicts Circuit Spatio-Temporal Filtering Properties , 2009, Front. Cell. Neurosci..

[33]  Giovanni Danese,et al.  Exploiting multi-core and many-core architectures for efficient simulation of biologically realistic models of Golgi cells , 2019, J. Parallel Distributed Comput..

[34]  Egidio D'Angelo,et al.  Combinatorial responses controlled by synaptic inhibition in the cerebellum granular layer. , 2010, Journal of neurophysiology.

[35]  S. Dieudonné,et al.  Submillisecond kinetics and low efficacy of parallel fibre‐Golgi cell synaptic currents in the rat cerebellum , 1998, The Journal of physiology.

[36]  R. Angus Silver,et al.  neuroConstruct: A Tool for Modeling Networks of Neurons in 3D Space , 2007, Neuron.

[37]  Kanichay Rt,et al.  Synaptic and Cellular Properties of the Feedforward Inhibitory Circuit within the Input Layer of the Cerebellar Cortex , 2008 .