GPUs Outperform Current HPC and Neuromorphic Solutions in Terms of Speed and Energy When Simulating a Highly-Connected Cortical Model

Efficient SNN multi-cores MAC array acceleration on SpiNNaker 2

The potential low-energy feature of the spiking neural network (SNN) engages the attention of the AI community. Only CPU-involved SNN processing inevitably results in an inherently long temporal span in the cases of large models and massive datasets. This study introduces the MAC array, a parallel architecture on each processing element (PE) of SpiNNaker 2, into the computational process of SNN inference. Based on the work of single-core optimization algorithms, we investigate the parallel acceleration algorithms for collaborating with multi-core MAC arrays. The proposed Echelon Reorder model information densification algorithm, along with the adapted multi-core two-stage splitting and authorization deployment strategies, achieves efficient spatio-temporal load balancing and optimization performance. We evaluate the performance by benchmarking a wide range of constructed SNN models to research on the influence degree of different factors. We also benchmark with two actual SNN models (the gesture recognition model of the real-world application and balanced random cortex-like network from neuroscience) on the neuromorphic multi-core hardware SpiNNaker 2. The echelon optimization algorithm with mixed processors realizes 74.28% and 85.78% memory footprint of the original MAC calculation on these two models, respectively. The execution time of echelon algorithms using only MAC or mixed processors accounts for ≤ 24.56% of the serial ARM baseline. Accelerating SNN inference with algorithms in this study is essentially the general sparse matrix-matrix multiplication (SpGEMM) problem. This article explicitly expands the application field of the SpGEMM issue to SNN, developing novel SpGEMM optimization algorithms fitting the SNN feature and MAC array.

A scalable implementation of the recursive least-squares algorithm for training spiking neural networks

Training spiking recurrent neural networks on neuronal recordings or behavioral tasks has become a popular way to study computations performed by the nervous system. As the size and complexity of neural recordings increase, there is a need for efficient algorithms that can train models in a short period of time using minimal resources. We present optimized CPU and GPU implementations of the recursive least-squares algorithm in spiking neural networks. The GPU implementation can train networks of one million neurons, with 100 million plastic synapses and a billion static synapses, about 1,000 times faster than an unoptimized reference CPU implementation. We demonstrate the code's utility by training a network, in less than an hour, to reproduce the activity of > 66, 000 recorded neurons of a mouse performing a decision-making task. The fast implementation enables a more interactive in-silico study of the dynamics and connectivity underlying multi-area computations. It also admits the possibility to train models as in-vivo experiments are being conducted, thus closing the loop between modeling and experiments.

Listen to the Brain-Auditory Sound Source Localization in Neuromorphic Computing Architectures

Conventional processing of sensory input often relies on uniform sampling leading to redundant information and unnecessary resource consumption throughout the entire processing pipeline. Neuromorphic computing challenges these conventions by mimicking biology and employing distributed event-based hardware. Based on the task of lateral auditory sound source localization (SSL), we propose a generic approach to map biologically inspired neural networks to neuromorphic hardware. First, we model the neural mechanisms of SSL based on the interaural level difference (ILD). Afterward, we identify generic computational motifs within the model and transform them into spike-based components. A hardware-specific step then implements them on neuromorphic hardware. We exemplify our approach by mapping the neural SSL model onto two platforms, namely the IBM TrueNorth Neurosynaptic System and SpiNNaker. Both implementations have been tested on synthetic and real-world data in terms of neural tunings and readout characteristics. For synthetic stimuli, both implementations provide a perfect readout (100% accuracy). Preliminary real-world experiments yield accuracies of 78% (TrueNorth) and 13% (SpiNNaker), RMSEs of 41∘ and 39∘, and MAEs of 18∘ and 29∘, respectively. Overall, the proposed mapping approach allows for the successful implementation of the same SSL model on two different neuromorphic architectures paving the way toward more hardware-independent neural SSL.

Geosmin suppresses defensive behaviour and elicits unusual neural responses in honey bees

Geosmin is an odorant produced by bacteria in moist soil. It has been found to be extraordinarily relevant to some insects, but the reasons for this are not yet fully understood. Here we report the first tests of the effect of geosmin on honey bees. A stinging assay showed that the defensive behaviour elicited by the bee's alarm pheromone component isoamyl acetate (IAA) is strongly suppressed by geosmin. Surprisingly, the suppression is, however, only present at very low geosmin concentrations, and disappears at higher concentrations. We investigated the underlying mechanisms at the level of the olfactory receptor neurons by means of electroantennography, finding the responses to mixtures of geosmin and IAA to be lower than to pure IAA, suggesting an interaction of both compounds at the olfactory receptor level. Calcium imaging of the antennal lobe (AL) revealed that neuronal responses to geosmin decreased with increasing concentration, correlating well with the observed behaviour. Computational modelling of odour transduction and coding in the AL suggests that a broader activation of olfactory receptor types by geosmin in combination with lateral inhibition could lead to the observed non-monotonic increasing-decreasing responses to geosmin and thus underlie the specificity of the behavioural response to low geosmin concentrations.

Efficient parameter calibration and real-time simulation of large-scale spiking neural networks with GeNN and NEST

Spiking neural networks (SNNs) represent the state-of-the-art approach to the biologically realistic modeling of nervous system function. The systematic calibration for multiple free model parameters is necessary to achieve robust network function and demands high computing power and large memory resources. Special requirements arise from closed-loop model simulation in virtual environments and from real-time simulation in robotic application. Here, we compare two complementary approaches to efficient large-scale and real-time SNN simulation. The widely used NEural Simulation Tool (NEST) parallelizes simulation across multiple CPU cores. The GPU-enhanced Neural Network (GeNN) simulator uses the highly parallel GPU-based architecture to gain simulation speed. We quantify fixed and variable simulation costs on single machines with different hardware configurations. As a benchmark model, we use a spiking cortical attractor network with a topology of densely connected excitatory and inhibitory neuron clusters with homogeneous or distributed synaptic time constants and in comparison to the random balanced network. We show that simulation time scales linearly with the simulated biological model time and, for large networks, approximately linearly with the model size as dominated by the number of synaptic connections. Additional fixed costs with GeNN are almost independent of model size, while fixed costs with NEST increase linearly with model size. We demonstrate how GeNN can be used for simulating networks with up to 3.5 · 10⁶ neurons (> 3 · 10¹²synapses) on a high-end GPU, and up to 250, 000 neurons (25 · 10⁹ synapses) on a low-cost GPU. Real-time simulation was achieved for networks with 100, 000 neurons. Network calibration and parameter grid search can be efficiently achieved using batch processing. We discuss the advantages and disadvantages of both approaches for different use cases.

neuroAIx-Framework: design of future neuroscience simulation systems exhibiting execution of the cortical microcircuit model 20× faster than biological real-time

Introduction

Research in the field of computational neuroscience relies on highly capable simulation platforms. With real-time capabilities surpassed for established models like the cortical microcircuit, it is time to conceive next-generation systems: neuroscience simulators providing significant acceleration, even for larger networks with natural density, biologically plausible multi-compartment models and the modeling of long-term and structural plasticity.

Methods

Stressing the need for agility to adapt to new concepts or findings in the domain of neuroscience, we have developed the neuroAIx-Framework consisting of an empirical modeling tool, a virtual prototype, and a cluster of FPGA boards. This framework is designed to support and accelerate the continuous development of such platforms driven by new insights in neuroscience.

Results

Based on design space explorations using this framework, we devised and realized an FPGA cluster consisting of 35 NetFPGA SUME boards.

Discussion

This system functions as an evaluation platform for our framework. At the same time, it resulted in a fully deterministic neuroscience simulation system surpassing the state of the art in both performance and energy efficiency. It is capable of simulating the microcircuit with 20× acceleration compared to biological real-time and achieves an energy efficiency of 48nJ per synaptic event.

Connectivity concepts in neuronal network modeling

Sustainable research on computational models of neuronal networks requires published models to be understandable, reproducible, and extendable. Missing details or ambiguities about mathematical concepts and assumptions, algorithmic implementations, or parameterizations hinder progress. Such flaws are unfortunately frequent and one reason is a lack of readily applicable standards and tools for model description. Our work aims to advance complete and concise descriptions of network connectivity but also to guide the implementation of connection routines in simulation software and neuromorphic hardware systems. We first review models made available by the computational neuroscience community in the repositories ModelDB and Open Source Brain, and investigate the corresponding connectivity structures and their descriptions in both manuscript and code. The review comprises the connectivity of networks with diverse levels of neuroanatomical detail and exposes how connectivity is abstracted in existing description languages and simulator interfaces. We find that a substantial proportion of the published descriptions of connectivity is ambiguous. Based on this review, we derive a set of connectivity concepts for deterministically and probabilistically connected networks and also address networks embedded in metric space. Beside these mathematical and textual guidelines, we propose a unified graphical notation for network diagrams to facilitate an intuitive understanding of network properties. Examples of representative network models demonstrate the practical use of the ideas. We hope that the proposed standardizations will contribute to unambiguous descriptions and reproducible implementations of neuronal network connectivity in computational neuroscience.

Fast Simulation of a Multi-Area Spiking Network Model of Macaque Cortex on an MPI-GPU Cluster

Spiking neural network models are increasingly establishing themselves as an effective tool for simulating the dynamics of neuronal populations and for understanding the relationship between these dynamics and brain function. Furthermore, the continuous development of parallel computing technologies and the growing availability of computational resources are leading to an era of large-scale simulations capable of describing regions of the brain of ever larger dimensions at increasing detail. Recently, the possibility to use MPI-based parallel codes on GPU-equipped clusters to run such complex simulations has emerged, opening up novel paths to further speed-ups. NEST GPU is a GPU library written in CUDA-C/C++ for large-scale simulations of spiking neural networks, which was recently extended with a novel algorithm for remote spike communication through MPI on a GPU cluster. In this work we evaluate its performance on the simulation of a multi-area model of macaque vision-related cortex, made up of about 4 million neurons and 24 billion synapses and representing 32 mm² surface area of the macaque cortex. The outcome of the simulations is compared against that obtained using the well-known CPU-based spiking neural network simulator NEST on a high-performance computing cluster. The results show not only an optimal match with the NEST statistical measures of the neural activity in terms of three informative distributions, but also remarkable achievements in terms of simulation time per second of biological activity. Indeed, NEST GPU was able to simulate a second of biological time of the full-scale macaque cortex model in its metastable state 3.1× faster than NEST using 32 compute nodes equipped with an NVIDIA V100 GPU each. Using the same configuration, the ground state of the full-scale macaque cortex model was simulated 2.4× faster than NEST.

Scaling and Benchmarking an Evolutionary Algorithm for Constructing Biophysical Neuronal Models

Single neuron models are fundamental for computational modeling of the brain's neuronal networks, and understanding how ion channel dynamics mediate neural function. A challenge in defining such models is determining biophysically realistic channel distributions. Here, we present an efficient, highly parallel evolutionary algorithm for developing such models, named NeuroGPU-EA. NeuroGPU-EA uses CPUs and GPUs concurrently to simulate and evaluate neuron membrane potentials with respect to multiple stimuli. We demonstrate a logarithmic cost for scaling the stimuli used in the fitting procedure. NeuroGPU-EA outperforms the typically used CPU based evolutionary algorithm by a factor of 10 on a series of scaling benchmarks. We report observed performance bottlenecks and propose mitigation strategies. Finally, we also discuss the potential of this method for efficient simulation and evaluation of electrophysiological waveforms.

NNMT: Mean-Field Based Analysis Tools for Neuronal Network Models

Mean-field theory of neuronal networks has led to numerous advances in our analytical and intuitive understanding of their dynamics during the past decades. In order to make mean-field based analysis tools more accessible, we implemented an extensible, easy-to-use open-source Python toolbox that collects a variety of mean-field methods for the leaky integrate-and-fire neuron model. The Neuronal Network Mean-field Toolbox (NNMT) in its current state allows for estimating properties of large neuronal networks, such as firing rates, power spectra, and dynamical stability in mean-field and linear response approximation, without running simulations. In this article, we describe how the toolbox is implemented, show how it is used to reproduce results of previous studies, and discuss different use-cases, such as parameter space explorations, or mapping different network models. Although the initial version of the toolbox focuses on methods for leaky integrate-and-fire neurons, its structure is designed to be open and extensible. It aims to provide a platform for collecting analytical methods for neuronal network model analysis, such that the neuroscientific community can take maximal advantage of them.

Photonic spiking neural networks with event-driven femtojoule optoelectronic neurons based on Izhikevich-inspired model

Photonic spiking neural networks (PSNNs) potentially offer exceptionally high throughput and energy efficiency compared to their electronic neuromorphic counterparts while maintaining their benefits in terms of event-driven computing capability. While state-of-the-art PSNN designs require a continuous laser pump, this paper presents a monolithic optoelectronic PSNN hardware design consisting of an MZI mesh incoherent network and event-driven laser spiking neurons. We designed, prototyped, and experimentally demonstrated this event-driven neuron inspired by the Izhikevich model incorporating both excitatory and inhibitory optical spiking inputs and producing optical spiking outputs accordingly. The optoelectronic neurons consist of two photodetectors for excitatory and inhibitory optical spiking inputs, electrical transistors' circuits providing spiking nonlinearity, and a laser for optical spiking outputs. Additional inclusion of capacitors and resistors complete the Izhikevich-inspired optoelectronic neurons, which receive excitatory and inhibitory optical spikes as inputs from other optoelectronic neurons. We developed a detailed optoelectronic neuron model in Verilog-A and simulated the circuit-level operation of various cases with excitatory input and inhibitory input signals. The experimental results closely resemble the simulated results and demonstrate how the excitatory inputs trigger the optical spiking outputs while the inhibitory inputs suppress the outputs. The nanoscale neuron designed in our monolithic PSNN utilizes quantum impedance conversion. It shows that estimated 21.09 fJ/spike input can trigger the output from on-chip nanolasers running at a maximum of 10 Gspike/second in the neural network. Utilizing the simulated neuron model, we conducted simulations on MNIST handwritten digits recognition using fully connected (FC) and convolutional neural networks (CNN). The simulation results show 90% accuracy on unsupervised learning and 97% accuracy on a supervised modified FC neural network. The benchmark shows our PSNN can achieve 50 TOP/J energy efficiency, which corresponds to 100 × throughputs and 1000 × energy-efficiency improvements compared to state-of-art electrical neuromorphic hardware such as Loihi and NeuroGrid.

A Scalable Approach to Modeling on Accelerated Neuromorphic Hardware

Neuromorphic systems open up opportunities to enlarge the explorative space for computational research. However, it is often challenging to unite efficiency and usability. This work presents the software aspects of this endeavor for the BrainScaleS-2 system, a hybrid accelerated neuromorphic hardware architecture based on physical modeling. We introduce key aspects of the BrainScaleS-2 Operating System: experiment workflow, API layering, software design, and platform operation. We present use cases to discuss and derive requirements for the software and showcase the implementation. The focus lies on novel system and software features such as multi-compartmental neurons, fast re-configuration for hardware-in-the-loop training, applications for the embedded processors, the non-spiking operation mode, interactive platform access, and sustainable hardware/software co-development. Finally, we discuss further developments in terms of hardware scale-up, system usability, and efficiency.

A Modular Workflow for Performance Benchmarking of Neuronal Network Simulations

Modern computational neuroscience strives to develop complex network models to explain dynamics and function of brains in health and disease. This process goes hand in hand with advancements in the theory of neuronal networks and increasing availability of detailed anatomical data on brain connectivity. Large-scale models that study interactions between multiple brain areas with intricate connectivity and investigate phenomena on long time scales such as system-level learning require progress in simulation speed. The corresponding development of state-of-the-art simulation engines relies on information provided by benchmark simulations which assess the time-to-solution for scientifically relevant, complementary network models using various combinations of hardware and software revisions. However, maintaining comparability of benchmark results is difficult due to a lack of standardized specifications for measuring the scaling performance of simulators on high-performance computing (HPC) systems. Motivated by the challenging complexity of benchmarking, we define a generic workflow that decomposes the endeavor into unique segments consisting of separate modules. As a reference implementation for the conceptual workflow, we develop beNNch: an open-source software framework for the configuration, execution, and analysis of benchmarks for neuronal network simulations. The framework records benchmarking data and metadata in a unified way to foster reproducibility. For illustration, we measure the performance of various versions of the NEST simulator across network models with different levels of complexity on a contemporary HPC system, demonstrating how performance bottlenecks can be identified, ultimately guiding the development toward more efficient simulation technology.

Simulating the Cortical Microcircuit Significantly Faster Than Real Time on the IBM INC-3000 Neural Supercomputer

This article employs the new IBM INC-3000 prototype FPGA-based neural supercomputer to implement a widely used model of the cortical microcircuit. With approximately 80,000 neurons and 300 Million synapses this model has become a benchmark network for comparing simulation architectures with regard to performance. To the best of our knowledge, the achieved speed-up factor is 2.4 times larger than the highest speed-up factor reported in the literature and four times larger than biological real time demonstrating the potential of FPGA systems for neural modeling. The work was performed at Jülich Research Centre in Germany and the INC-3000 was built at the IBM Almaden Research Center in San Jose, CA, United States. For the simulation of the microcircuit only the programmable logic part of the FPGA nodes are used. All arithmetic is implemented with single-floating point precision. The original microcircuit network with linear LIF neurons and current-based exponential-decay-, alpha-function- as well as beta-function-shaped synapses was simulated using exact exponential integration as ODE solver method. In order to demonstrate the flexibility of the approach, additionally networks with non-linear neuron models (AdEx, Izhikevich) and conductance-based synapses were simulated, applying Runge–Kutta and Parker–Sochacki solver methods. In all cases, the simulation-time speed-up factor did not decrease by more than a very few percent. It finally turns out that the speed-up factor is essentially limited by the latency of the INC-3000 communication system.

Opportunities for neuromorphic computing algorithms and applications

Neuromorphic computing technologies will be important for the future of computing, but much of the work in neuromorphic computing has focused on hardware development. Here, we review recent results in neuromorphic computing algorithms and applications. We highlight characteristics of neuromorphic computing technologies that make them attractive for the future of computing and we discuss opportunities for future development of algorithms and applications on these systems.

Dynamical Characteristics of Recurrent Neuronal Networks Are Robust Against Low Synaptic Weight Resolution

The representation of the natural-density, heterogeneous connectivity of neuronal network models at relevant spatial scales remains a challenge for Computational Neuroscience and Neuromorphic Computing. In particular, the memory demands imposed by the vast number of synapses in brain-scale network simulations constitute a major obstacle. Limiting the number resolution of synaptic weights appears to be a natural strategy to reduce memory and compute load. In this study, we investigate the effects of a limited synaptic-weight resolution on the dynamics of recurrent spiking neuronal networks resembling local cortical circuits and develop strategies for minimizing deviations from the dynamics of networks with high-resolution synaptic weights. We mimic the effect of a limited synaptic weight resolution by replacing normally distributed synaptic weights with weights drawn from a discrete distribution, and compare the resulting statistics characterizing firing rates, spike-train irregularity, and correlation coefficients with the reference solution. We show that a naive discretization of synaptic weights generally leads to a distortion of the spike-train statistics. If the weights are discretized such that the mean and the variance of the total synaptic input currents are preserved, the firing statistics remain unaffected for the types of networks considered in this study. For networks with sufficiently heterogeneous in-degrees, the firing statistics can be preserved even if all synaptic weights are replaced by the mean of the weight distribution. We conclude that even for simple networks with non-plastic neurons and synapses, a discretization of synaptic weights can lead to substantial deviations in the firing statistics unless the discretization is performed with care and guided by a rigorous validation process. For the network model used in this study, the synaptic weights can be replaced by low-resolution weights without affecting its macroscopic dynamical characteristics, thereby saving substantial amounts of memory.

Benchmarking Highly Parallel Hardware for Spiking Neural Networks in Robotics

Animal brains still outperform even the most performant machines with significantly lower speed. Nonetheless, impressive progress has been made in robotics in the areas of vision, motion- and path planning in the last decades. Brain-inspired Spiking Neural Networks (SNN) and the parallel hardware necessary to exploit their full potential have promising features for robotic application. Besides the most obvious platform for deploying SNN, brain-inspired neuromorphic hardware, Graphical Processing Units (GPU) are well capable of parallel computing as well. Libraries for generating CUDA-optimized code, like GeNN and affordable embedded systems make them an attractive alternative due to their low price and availability. While a few performance tests exist, there has been a lack of benchmarks targeting robotic applications. We compare the performance of a neural Wavefront algorithm as a representative of use cases in robotics on different hardware suitable for running SNN simulations. The SNN used for this benchmark is modeled in the simulator-independent declarative language PyNN, which allows using the same model for different simulator backends. Our emphasis is the comparison between Nest, running on serial CPU, SpiNNaker, as a representative of neuromorphic hardware, and an implementation in GeNN. Beyond that, we also investigate the differences of GeNN deployed to different hardware. A comparison between the different simulators and hardware is performed with regard to total simulation time, average energy consumption per run, and the length of the resulting path. We hope that the insights gained about performance details of parallel hardware solutions contribute to developing more efficient SNN implementations for robotics.

PyGeNN: A Python Library for GPU-Enhanced Neural Networks

More than half of the Top 10 supercomputing sites worldwide use GPU accelerators and they are becoming ubiquitous in workstations and edge computing devices. GeNN is a C++ library for generating efficient spiking neural network simulation code for GPUs. However, until now, the full flexibility of GeNN could only be harnessed by writing model descriptions and simulation code in C++. Here we present PyGeNN, a Python package which exposes all of GeNN's functionality to Python with minimal overhead. This provides an alternative, arguably more user-friendly, way of using GeNN and allows modelers to use GeNN within the growing Python-based machine learning and computational neuroscience ecosystems. In addition, we demonstrate that, in both Python and C++ GeNN simulations, the overheads of recording spiking data can strongly affect runtimes and show how a new spike recording system can reduce these overheads by up to 10×. Using the new recording system, we demonstrate that by using PyGeNN on a modern GPU, we can simulate a full-scale model of a cortical column faster even than real-time neuromorphic systems. Finally, we show that long simulations of a smaller model with complex stimuli and a custom three-factor learning rule defined in PyGeNN can be simulated almost two orders of magnitude faster than real-time.

Fast Simulations of Highly-Connected Spiking Cortical Models Using GPUs

Over the past decade there has been a growing interest in the development of parallel hardware systems for simulating large-scale networks of spiking neurons. Compared to other highly-parallel systems, GPU-accelerated solutions have the advantage of a relatively low cost and a great versatility, thanks also to the possibility of using the CUDA-C/C++ programming languages. NeuronGPU is a GPU library for large-scale simulations of spiking neural network models, written in the C++ and CUDA-C++ programming languages, based on a novel spike-delivery algorithm. This library includes simple LIF (leaky-integrate-and-fire) neuron models as well as several multisynapse AdEx (adaptive-exponential-integrate-and-fire) neuron models with current or conductance based synapses, different types of spike generators, tools for recording spikes, state variables and parameters, and it supports user-definable models. The numerical solution of the differential equations of the dynamics of the AdEx models is performed through a parallel implementation, written in CUDA-C++, of the fifth-order Runge-Kutta method with adaptive step-size control. In this work we evaluate the performance of this library on the simulation of a cortical microcircuit model, based on LIF neurons and current-based synapses, and on balanced networks of excitatory and inhibitory neurons, using AdEx or Izhikevich neuron models and conductance-based or current-based synapses. On these models, we will show that the proposed library achieves state-of-the-art performance in terms of simulation time per second of biological activity. In particular, using a single NVIDIA GeForce RTX 2080 Ti GPU board, the full-scale cortical-microcircuit model, which includes about 77,000 neurons and 3 · 10⁸ connections, can be simulated at a speed very close to real time, while the simulation time of a balanced network of 1,000,000 AdEx neurons with 1,000 connections per neuron was about 70 s per second of biological activity.

Larger GPU-accelerated brain simulations with procedural connectivity

Simulations are an important tool for investigating brain function but large models are needed to faithfully reproduce the statistics and dynamics of brain activity. Simulating large spiking neural network models has, until now, needed so much memory for storing synaptic connections that it required high performance computer systems. Here, we present an alternative simulation method we call 'procedural connectivity' where connectivity and synaptic weights are generated 'on the fly' instead of stored and retrieved from memory. This method is particularly well suited for use on graphical processing units (GPUs)-which are a common fixture in many workstations. Using procedural connectivity and an additional GPU code generation optimization, we can simulate a recent model of the macaque visual cortex with 4.13 × 106 neurons and 24.2 × 109 synapses on a single GPU-a significant step forward in making large-scale brain modeling accessible to more researchers.

Understanding Computational Costs of Cellular-Level Brain Tissue Simulations Through Analytical Performance Models

Computational modeling and simulation have become essential tools in the quest to better understand the brain's makeup and to decipher the causal interrelations of its components. The breadth of biochemical and biophysical processes and structures in the brain has led to the development of a large variety of model abstractions and specialized tools, often times requiring high performance computing resources for their timely execution. What has been missing so far was an in-depth analysis of the complexity of the computational kernels, hindering a systematic approach to identifying bottlenecks of algorithms and hardware. If whole brain models are to be achieved on emerging computer generations, models and simulation engines will have to be carefully co-designed for the intrinsic hardware tradeoffs. For the first time, we present a systematic exploration based on analytic performance modeling. We base our analysis on three in silico models, chosen as representative examples of the most widely employed modeling abstractions: current-based point neurons, conductance-based point neurons and conductance-based detailed neurons. We identify that the synaptic modeling formalism, i.e. current or conductance-based representation, and not the level of morphological detail, is the most significant factor in determining the properties of memory bandwidth saturation and shared-memory scaling of in silico models. Even though general purpose computing has, until now, largely been able to deliver high performance, we find that for all types of abstractions, network latency and memory bandwidth will become severe bottlenecks as the number of neurons to be simulated grows. By adapting and extending a performance modeling approach, we deliver a first characterization of the performance landscape of brain tissue simulations, allowing us to pinpoint current bottlenecks for state-of-the-art in silico models, and make projections for future hardware and software requirements.

Real-time cortical simulation on neuromorphic hardware

Real-time simulation of a large-scale biologically representative spiking neural network is presented, through the use of a heterogeneous parallelization scheme and SpiNNaker neuromorphic hardware. A published cortical microcircuit model is used as a benchmark test case, representing ≈1 mm² of early sensory cortex, containing 77 k neurons and 0.3 billion synapses. This is the first hard real-time simulation of this model, with 10 s of biological simulation time executed in 10 s wall-clock time. This surpasses best-published efforts on HPC neural simulators (3 × slowdown) and GPUs running optimized spiking neural network (SNN) libraries (2 × slowdown). Furthermore, the presented approach indicates that real-time processing can be maintained with increasing SNN size, breaking the communication barrier incurred by traditional computing machinery. Model results are compared to an established HPC simulator baseline to verify simulation correctness, comparing well across a range of statistical measures. Energy to solution and energy per synaptic event are also reported, demonstrating that the relatively low-tech SpiNNaker processors achieve a 10 × reduction in energy relative to modern HPC systems, and comparable energy consumption to modern GPUs. Finally, system robustness is demonstrated through multiple 12 h simulations of the cortical microcircuit, each simulating 12 h of biological time, and demonstrating the potential of neuromorphic hardware as a neuroscience research tool for studying complex spiking neural networks over extended time periods. This article is part of the theme issue 'Harmonizing energy-autonomous computing and intelligence'.

Brian2GeNN: accelerating spiking neural network simulations with graphics hardware

“Brian” is a popular Python-based simulator for spiking neural networks, commonly used in computational neuroscience. GeNN is a C++-based meta-compiler for accelerating spiking neural network simulations using consumer or high performance grade graphics processing units (GPUs). Here we introduce a new software package, Brian2GeNN, that connects the two systems so that users can make use of GeNN GPU acceleration when developing their models in Brian, without requiring any technical knowledge about GPUs, C++ or GeNN. The new Brian2GeNN software uses a pipeline of code generation to translate Brian scripts into C++ code that can be used as input to GeNN, and subsequently can be run on suitable NVIDIA GPU accelerators. From the user’s perspective, the entire pipeline is invoked by adding two simple lines to their Brian scripts. We have shown that using Brian2GeNN, two non-trivial models from the literature can run tens to hundreds of times faster than on CPU.

An unsupervised neuromorphic clustering algorithm

Brains perform complex tasks using a fraction of the power that would be required to do the same on a conventional computer. New neuromorphic hardware systems are now becoming widely available that are intended to emulate the more power efficient, highly parallel operation of brains. However, to use these systems in applications, we need "neuromorphic algorithms" that can run on them. Here we develop a spiking neural network model for neuromorphic hardware that uses spike timing-dependent plasticity and lateral inhibition to perform unsupervised clustering. With this model, time-invariant, rate-coded datasets can be mapped into a feature space with a specified resolution, i.e., number of clusters, using exclusively neuromorphic hardware. We developed and tested implementations on the SpiNNaker neuromorphic system and on GPUs using the GeNN framework. We show that our neuromorphic clustering algorithm achieves results comparable to those of conventional clustering algorithms such as self-organizing maps, neural gas or k-means clustering. We then combine it with a previously reported supervised neuromorphic classifier network to demonstrate its practical use as a neuromorphic preprocessing module.

Efficient Reward-Based Structural Plasticity on a SpiNNaker 2 Prototype

Advances in neuroscience uncover the mechanisms employed by the brain to efficiently solve complex learning tasks with very limited resources. However, the efficiency is often lost when one tries to port these findings to a silicon substrate, since brain-inspired algorithms often make extensive use of complex functions, such as random number generators, that are expensive to compute on standard general purpose hardware. The prototype chip of the second generation SpiNNaker system is designed to overcome this problem. Low-power advanced RISC machine (ARM) processors equipped with a random number generator and an exponential function accelerator enable the efficient execution of brain-inspired algorithms. We implement the recently introduced reward-based synaptic sampling model that employs structural plasticity to learn a function or task. The numerical simulation of the model requires to update the synapse variables in each time step including an explorative random term. To the best of our knowledge, this is the most complex synapse model implemented so far on the SpiNNaker system. By making efficient use of the hardware accelerators and numerical optimizations, the computation time of one plasticity update is reduced by a factor of 2. This, combined with fitting the model into to the local static random access memory (SRAM), leads to 62% energy reduction compared to the case without accelerators and the use of external dynamic random access memory (DRAM). The model implementation is integrated into the SpiNNaker software framework allowing for scalability onto larger systems. The hardware-software system presented in this paper paves the way for power-efficient mobile and biomedical applications with biologically plausible brain-inspired algorithms.