Strategies in SIMD Computing for Complex Neural Bioinspired Applications

2009 NASA/ESA Conference on Adaptive Hardware and Systems Pub Date : 2009-07-29 DOI:10.1109/AHS.2009.31

J. Madrenas, J. Moreno

引用次数: 10

Abstract

The scalable architecture of a multiprocessor intended to efficiently accelerate the emulation of large-scale complex systems, in particular massively-parallel bioinspired neural networks, is introduced in this paper. In order to cope with the required processing complexity of a target network size of large number of neurons and synapses, the SIMD configuration is adopted. Special flow-control instructions are proposed to support conditional execution. Also, a modified digital AER scheme allows for the compact emulation of interconnects. Due to its programmable characteristics, the architecture is flexible enough to support the emulation of different neural models and other homogeneous parallel applications.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

复杂神经生物应用SIMD计算策略

本文介绍了一种可扩展的多处理器体系结构，旨在有效地加速大规模复杂系统的仿真，特别是大规模并行生物神经网络的仿真。为了应对具有大量神经元和突触的目标网络规模所要求的处理复杂度，采用了SIMD配置。提出了特殊的流控制指令来支持条件执行。此外，改进的数字AER方案允许对互连进行紧凑的仿真。由于其可编程特性，该架构足够灵活，可以支持不同神经模型和其他同构并行应用的仿真。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

2009 NASA/ESA Conference on Adaptive Hardware and Systems

自引率

0.00%

发文量

期刊最新文献

Dynamically Adapted Low-Energy Fault Tolerant Processors Partial Bitstream 2-D Core Relocation for Reconfigurable Architectures Automated Wire Antennas Design using Dynamic Dominance Evolutionary Algorithm Adaptive Hardware Real-Time Task Scheduler of Multi-Core ATPA Environment Evolutionary Algorithms in Unreliable Memory