Using Neuromorphic Hardware for the Scalable Execution of Massively Parallel, Communication-Intensive Algorithms
Using Neuromorphic Hardware for the Scalable Execution of Massively Parallel, Communication-Intensive Algorithms
复制标题
DOI：
10.1109/ucc-companion.2018.00040
复制
发表时间：
2018-12
期刊：
影响因子：
--
通讯作者：
Louis Blin;Ahsan Javed Awan;T. Heinis
中科院分区：
文献类型：
--
作者：
Louis Blin;Ahsan Javed Awan;T. Heinis
像SpiNNaker这样的神经形态硬件提供了大量的并行性和小有效载荷的有效通信，以加速神经网络中尖峰神经元的模拟。在本文中，我们证明了这种硬件也有利于其他应用程序，需要大量的并行性和大规模的小消息交换。更具体地说，我们研究了SpiNNaker上的PageRank的可扩展性，并将其与传统硬件上的实现进行比较。在我们的实验中，我们表明SpiNNaker上的PageRank比传统的多核架构更好。
Neuromorphic hardware like SpiNNaker offers massive parallelism and efficient communication of small payloads to accelerate the simulation of spiking neurons in neural networks. In this paper, we demonstrate that this hardware is also beneficial for other for applications which require massive parallelism and the large-scale exchange of small messages. More specifically, we study the scalability of PageRank on SpiNNaker and compare it to an implementation on traditional hardware. In our experiments, we show that PageRank on SpiNNaker scales better than on traditional multicore architectures.