Power analysis of large-scale, real-time neural networks on SpiNNaker

Stromatias, Evangelos; Galluppi, Francesco; Patterson, Cameron; Furber, Steve

doi:10.1109/ijcnn.2013.6706927

Cited by 60 publications

(59 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We introduce a classification of computing systems based on neuroprocessors, according to types of associations among their elements [4][5][6][7].…”

Section: Mathematical Formalization Of Data Processing In Robotic Neumentioning

confidence: 99%

See 1 more Smart Citation

Mathematical support and software for data processing in robotic neurocomputer systems

Romanchuk

2018

MATEC Web Conf.

View full text Add to dashboard Cite

show abstract

“…We introduce a classification of computing systems based on neuroprocessors, according to types of associations among their elements [4][5][6][7].…”

Section: Mathematical Formalization Of Data Processing In Robotic Neumentioning

confidence: 99%

“…-P. Kolinummo, P. PASI Pulkkinen, T. Hämäläinen and J. Saarinen [3], E. Stromatias, F. Galluppi, C. Patterson, S. Furber [5], K. Esseret [6] have chosen a narrow field of research: parallel execution of self-organizing maps (Kohonen maps) on a neurocomputer (emulator).…”

Section: Introductionmentioning

confidence: 99%

Mathematical support and software for data processing in robotic neurocomputer systems

Romanchuk

2018

MATEC Web Conf.

View full text Add to dashboard Cite

show abstract

“…Therefore, fast and powerefficient execution of spiking neural network models assumes a driving role, at the cross-road between embedded and highperformance computing, shaping the evolution of the architecture of specialized and general-purpose multicore/manycore systems. See, for example, the TrueNorth [41] low-power specialized hardware architecture for embedded applications and [42] about the power consumption of the SpiNNaker specialized hardware architecture, based on the combination of embedded multicores and a dedicated networking infrastructure. About the strategy based on more standard HPC platforms and general-purpose simulators, see, for example, [43,44].…”

Section: Neural Networkmentioning

confidence: 99%

Power-Efficient Computing: Experiences from the COSA Project

Cesini¹,

Corni²,

Falabella³

et al. 2017

Scientific Programming

View full text Add to dashboard Cite

Energy consumption is today one of the most relevant issues in operating HPC systems for scientific applications. The use of unconventional computing systems is therefore of great interest for several scientific communities looking for a better tradeoff between time-to-solution and energy-to-solution. In this context, the performance assessment of processors with a high ratio of performance per watt is necessary to understand how to realize energy-efficient computing systems for scientific applications, using this class of processors. Computing On SOC Architecture (COSA) is a three-year project (2015)(2016)(2017) funded by the Scientific Commission V of the Italian Institute for Nuclear Physics (INFN), which aims to investigate the performance and the total cost of ownership offered by computing systems based on commodity low-power Systems on Chip (SoCs) and high energyefficient systems based on GP-GPUs. In this work, we present the results of the project analyzing the performance of several scientific applications on several GPU-and SoC-based systems. We also describe the methodology we have used to measure energy performance and the tools we have implemented to monitor the power drained by applications while running.

show abstract

“…The project was originally conceived at the University of Manchester (APT Group, 2015) and is today further developed as part of the Human Brain Project. A SpiNNaker chip contains 18 ARM microprocessors, each of which is capable of simulating around 1000 spiking neurons with about 1000 synapses each in biological real-time (Painkras et al, 2012;Stromatias, Galluppi, Patterson, & Furber, 2013). The architecture allows for an integration of up to 65 536 chips .…”

Section: Spinnakermentioning

confidence: 99%

Neuromorphic implementations of neurobiological learning algorithms for spiking neural networks

2015

View full text Add to dashboard Cite

a b s t r a c tThe application of biologically inspired methods in design and control has a long tradition in robotics. Unlike previous approaches in this direction, the emerging field of neurorobotics not only mimics biological mechanisms at a relatively high level of abstraction but employs highly realistic simulations of actual biological nervous systems. Even today, carrying out these simulations efficiently at appropriate timescales is challenging. Neuromorphic chip designs specially tailored to this task therefore offer an interesting perspective for neurorobotics. Unlike Von Neumann CPUs, these chips cannot be simply programmed with a standard programming language. Like real brains, their functionality is determined by the structure of neural connectivity and synaptic efficacies. Enabling higher cognitive functions for neurorobotics consequently requires the application of neurobiological learning algorithms to adjust synaptic weights in a biologically plausible way. In this paper, we therefore investigate how to program neuromorphic chips by means of learning. First, we provide an overview over selected neuromorphic chip designs and analyze them in terms of neural computation, communication systems and software infrastructure. On the theoretical side, we review neurobiological learning techniques. Based on this overview, we then examine on-die implementations of these learning algorithms on the considered neuromorphic chips. A final discussion puts the findings of this work into context and highlights how neuromorphic hardware can potentially advance the field of autonomous robot systems. The paper thus gives an in-depth overview of neuromorphic implementations of basic mechanisms of synaptic plasticity which are required to realize advanced cognitive capabilities with spiking neural networks.

show abstract

Power analysis of large-scale, real-time neural networks on SpiNNaker

Cited by 60 publications

References 32 publications

Mathematical support and software for data processing in robotic neurocomputer systems

Mathematical support and software for data processing in robotic neurocomputer systems

Power-Efficient Computing: Experiences from the COSA Project

Neuromorphic implementations of neurobiological learning algorithms for spiking neural networks

Contact Info

Product

Resources

About