High performance computing on SpiNNaker neuromorphic platform: A case study for energy efficient image processing

Sugiarto, Indar; Liu, Gengting; Davidson, Simon; Plana, Luis A.; Furber, Steve

doi:10.1109/pccc.2016.7820645

Cited by 19 publications

(14 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, since SpiNNaker machines are built on top of a general purpose processor, it is still possible, using the direct programming approach on SpiNNaker, to implement a DL concept on SpiNNaker. This approach proves that SpiNNaker can also be used for general purpose computing, even in a domain far from spiking neural network, such as in image process-ing [57]. The following sub-sections provide examples of applications that are closely related to the concept of DL.…”

Section: Case Study With Spinnakermentioning

confidence: 86%

Understanding a Deep Learning Technique through a Neuromorphic System a Case Study with SpiNNaker Neuromorphic Platform

Sugiarto

Pasila

2018

MATEC Web Conf.

Self Cite

View full text Add to dashboard Cite

Abstract. Deep learning (DL) has been considered as a breakthrough technique in the field of artificial intelligence and machine learning. Conceptually, it relies on a many-layer network that exhibits a hierarchically non-linear processing capability. Some DL architectures such as deep neural networks, deep belief networks and recurrent neural networks have been developed and applied to many fields with incredible results, even comparable to human intelligence. However, many researchers are still sceptical about its true capability: can the intelligence demonstrated by deep learning technique be applied for general tasks? This question motivates the emergence of another research discipline: neuromorphic computing (NC). In NC, researchers try to identify the most fundamental ingredients that construct intelligence behaviour produced by the brain itself. To achieve this, neuromorphic systems are developed to mimic the brain functionality down to cellular level. In this paper, a neuromorphic platform called SpiNNaker is described and evaluated in order to understand its potential use as a platform for a deep learning approach. This paper is a literature review that contains comparative study on algorithms that have been implemented in SpiNNaker.

show abstract

Section: Case Study With Spinnakermentioning

confidence: 86%

Understanding a Deep Learning Technique through a Neuromorphic System a Case Study with SpiNNaker Neuromorphic Platform

Sugiarto

Pasila

2018

MATEC Web Conf.

Self Cite

View full text Add to dashboard Cite

show abstract

“…These identical copies can then use the shared-memory mechanism to harness hardware multi-threading. A similar mechanism has been used successfully to achieve high performance image processing on SpiNNaker as described in [8]. The zoomed-in representation of node P6 of the 9-node task graph illustrate this amelioration strategy.…”

Section: Task Mapping On a Spinnaker Machinementioning

confidence: 99%

Optimized task graph mapping on a many-core neuromorphic supercomputer

Sugiarto

Campos

Dahir

et al. 2017

2017 IEEE High Performance Extreme Computing Conference (HPEC)

Self Cite

View full text Add to dashboard Cite

Abstract-This paper presents an approach for improving the overall performance of a general purpose application running as a task graph on a many-core neuromorphic supercomputer. Our task graph framework is based on graceful degradation and amelioration paradigms that strive to achieve high reliability and performance by incorporating fault tolerance and task spawning features. The optimization is applied on an instance of the task graph by performing a soft load balancing on the data traffic between nodes in the graph. We implemented the framework and its optimization on SpiNNaker, a manycore neuromorphic platform containing a million ARM9 processing cores. We evaluate our method using several static mapping examples, where some of them were generated using an evolutionary algorithm. The experiment demonstrates that a performance improvement of up to 8.2% can be achieved when implementing our algorithm on a fully-utilized SpiNNaker communication infrastructure.

show abstract

“…In this field, Spiking Neural Networks (SNN) are used to form the computations. The SpiNNaker Project is one example [Sugiarto et al, 2016] and forms part of the Human Brain Project 19 . The Darwin Neural Processing Unit is another exciting example of an ASIC co-processer based on SNN [Shen et al, 2016].…”

Section: Asic Deep Learning Processorsmentioning

confidence: 99%

Custom Hardware Versus Cloud Computing in Big Data

Lightbody

Browne

Haberland

2017

Advanced Information and Knowledge Processing

View full text Add to dashboard Cite

The computational and data handling challenges in big data are immense yet a market is steadily growing traditionally supported by technologies such as Hadoop for management and processing of huge and unstructured datasets. With this ever increasing deluge of data we now need the algorithms, tools and computing infrastructure to handle the extremely computationally intense data analytics, looking for patterns and information pertinent to creating a market edge for a range of applications. Cloud computing has provided opportunities for scalable high-performance solutions without the initial outlay of developing and creating the core infrastructure. One vendor in particular, Amazon Web Services, has been leading this field. However, other solutions exist to take on the computational load of big data analytics. This chapter provides an overview of the extent of applications in which big data analytics is used. Then an overview is given of some of the high-performance computing options that are available, ranging from multiple Central Processing Unit (CPU) setups, Graphical Processing Units (GPUs), Field Programmable Gate Arrays (FPGAs) and cloud solutions. The chapter concludes by looking at some of the state of the art solutions for deep learning platforms in which custom hardware such as FPGAs and Application Specific Integrated Circuits (ASICs) are used within a cloud platform for key computational bottlenecks.

show abstract

High performance computing on SpiNNaker neuromorphic platform: A case study for energy efficient image processing

Cited by 19 publications

References 9 publications

Understanding a Deep Learning Technique through a Neuromorphic System a Case Study with SpiNNaker Neuromorphic Platform

Understanding a Deep Learning Technique through a Neuromorphic System a Case Study with SpiNNaker Neuromorphic Platform

Optimized task graph mapping on a many-core neuromorphic supercomputer

Custom Hardware Versus Cloud Computing in Big Data

Contact Info

Product

Resources

About