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

Heittmann, Arne; Psychou, Georgia; Trensch, Guido; Cox, Charles E.; Wilcke, W. W.; Diesmann, Markus; Noll, Tobias G.

doi:10.3389/fnins.2021.728460

Cited by 12 publications

(24 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To our best knowledge, we report the lowest RTF so far at a competitive energy consumption (table 2). There are, however, data [22] on an even smaller RTF for a dedicated FPGA supercomputer using on-the-fly generation of connectivity. Our results expose that cache sensitive binding of threads increases performance.…”

Section: Discussionmentioning

confidence: 99%

Sub-realtime simulation of a neuronal network of natural density

Kurth

Senk²,

Terhorst³

et al. 2022

Neuromorph. Comput. Eng.

Self Cite

View full text Add to dashboard Cite

Full scale simulations of neuronal network models of the brain are challenging due to the high density of connections between neurons. This contribution reports run times shorter than the simulated span of biological time for a full scale model of the local cortical microcircuit with explicit representation of synapses on a recent conventional compute node. Realtime performance is relevant for robotics and closed-loop applications while sub-realtime is desirable for the study of learning and development in the brain, processes extending over hours and days of biological time.

show abstract

Section: Discussionmentioning

confidence: 99%

Sub-realtime simulation of a neuronal network of natural density

Kurth

Senk²,

Terhorst³

et al. 2022

Neuromorph. Comput. Eng.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The first is a functional requirement: even though the current development does not yet include plasticity, in order to be able to cope with synaptic and structural plasticity algorithms in future, the synaptic connections must be stored, accessible, and changeable. In contrast, for static networks, performance-efficient solutions have been developed which makes use of a procedural connectivity generation approach (Knight and Nowotny, 2021 ; Heittmann et al, 2022 ) where the synaptic connections are determined algorithmically during the simulation, thus avoiding having to retrieve them from memory. The second is a resource constraint due to technical limitations of the technology: fast, low-latency, on-chip block RAM (BRAM) would be ideal to hold this data, but BRAM is a limited FPGA resource and the memory requirement for storing a network's connectivity data is demanding.…”

Section: Overview Of the Hybrid Neuromorphic Compute (Hnc) Nodementioning

confidence: 99%

“…The system is highly flexible and applications can off-load algorithms and accelerate them using the programmable logic of the Zynq SoC devices. An example of such an application is the implementation of the cortical microcircuit model (Potjans and Diesmann, 2014 ) on the INC-3000 presented in Heittmann et al ( 2022 )—a reproduction of an equivalent NEST implementation and on the SpiNNaker neuromorphic system (cf. van Albada et al, 2018 ).…”

Section: Introductionmentioning

confidence: 99%

A System-on-Chip Based Hybrid Neuromorphic Compute Node Architecture for Reproducible Hyper-Real-Time Simulations of Spiking Neural Networks

Trensch

Morrison

2022

Front. Neuroinform.

Self Cite

View full text Add to dashboard Cite

Despite the great strides neuroscience has made in recent decades, the underlying principles of brain function remain largely unknown. Advancing the field strongly depends on the ability to study large-scale neural networks and perform complex simulations. In this context, simulations in hyper-real-time are of high interest, as they would enable both comprehensive parameter scans and the study of slow processes, such as learning and long-term memory. Not even the fastest supercomputer available today is able to meet the challenge of accurate and reproducible simulation with hyper-real acceleration. The development of novel neuromorphic computer architectures holds out promise, but the high costs and long development cycles for application-specific hardware solutions makes it difficult to keep pace with the rapid developments in neuroscience. However, advances in System-on-Chip (SoC) device technology and tools are now providing interesting new design possibilities for application-specific implementations. Here, we present a novel hybrid software-hardware architecture approach for a neuromorphic compute node intended to work in a multi-node cluster configuration. The node design builds on the Xilinx Zynq-7000 SoC device architecture that combines a powerful programmable logic gate array (FPGA) and a dual-core ARM Cortex-A9 processor extension on a single chip. Our proposed architecture makes use of both and takes advantage of their tight coupling. We show that available SoC device technology can be used to build smaller neuromorphic computing clusters that enable hyper-real-time simulation of networks consisting of tens of thousands of neurons, and are thus capable of meeting the high demands for modeling and simulation in neuroscience.

show abstract

“…The system enables energy-efficient neuronal network simulations, offering highly accelerated operations. Another promising project in this field is SpiNNaker (Furber et al, 2014 ), which recently achieved biological real-time simulations of a cortical microcircuit model (Rhodes et al, 2020 ) proposed by Potjans and Diesmann ( 2014 ) (which has since been simulated sub-realtime with NEST (Kurth et al, 2022 ) and with an FPGA-based neural supercomputer (Heittmann et al, 2022 ). This result was made possible by its architecture designed for efficient spike communication, performed with an optimized transmission system of small data packets.…”

Section: Introductionmentioning

confidence: 99%

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

Tiddia

Golosio

Albers

et al. 2022

Front. Neuroinform.

View full text Add to dashboard Cite

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.

show abstract

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

Cited by 12 publications

References 43 publications

Sub-realtime simulation of a neuronal network of natural density

Sub-realtime simulation of a neuronal network of natural density

A System-on-Chip Based Hybrid Neuromorphic Compute Node Architecture for Reproducible Hyper-Real-Time Simulations of Spiking Neural Networks

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

Contact Info

Product

Resources

About