采用高性能数字神经处理器的multiineuro计算机系统

SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers Pub Date : 1995-07-26 DOI:10.1109/SICE.1995.526722

T. Saito, Y. Baba, K. Nakagawa, Y. Fuwa, Y. Shimokawa

{"title":"采用高性能数字神经处理器的multiineuro计算机系统","authors":"T. Saito, Y. Baba, K. Nakagawa, Y. Fuwa, Y. Shimokawa","doi":"10.1109/SICE.1995.526722","DOIUrl":null,"url":null,"abstract":"We have developed an accelerator device \"MULTINEURO-A\" which incorporates multiple processors (one master node and plural slave nodes) dedicated to neural network high-speed computing realized on an engineering workstation. Nodes are connected by two data paths, broadcast bus and ring bus. Processing speed is in proportion to the number of processors. It is built in two boards (master and slave) having 64 VLSI processors that offers to 1.5 billion connections per seconds. It processes forward/backward calculation of multilayer perceptron type neural networks, feedback type neural networks such as Hopfield model, and any other types by programming for parallel processing.","PeriodicalId":344374,"journal":{"name":"SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers","volume":"77 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1995-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A MULTINEURO computer system by using high performance digital neuro processor\",\"authors\":\"T. Saito, Y. Baba, K. Nakagawa, Y. Fuwa, Y. Shimokawa\",\"doi\":\"10.1109/SICE.1995.526722\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We have developed an accelerator device \\\"MULTINEURO-A\\\" which incorporates multiple processors (one master node and plural slave nodes) dedicated to neural network high-speed computing realized on an engineering workstation. Nodes are connected by two data paths, broadcast bus and ring bus. Processing speed is in proportion to the number of processors. It is built in two boards (master and slave) having 64 VLSI processors that offers to 1.5 billion connections per seconds. It processes forward/backward calculation of multilayer perceptron type neural networks, feedback type neural networks such as Hopfield model, and any other types by programming for parallel processing.\",\"PeriodicalId\":344374,\"journal\":{\"name\":\"SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers\",\"volume\":\"77 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1995-07-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/SICE.1995.526722\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SICE.1995.526722","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

我们开发了一种集成了多个处理器(一个主节点和多个从节点)的加速装置“multi - euroa”，用于在工程工作站上实现神经网络高速计算。节点之间通过广播总线和环形总线两条数据路径连接。处理速度与处理器的数量成正比。它内置在两块板(主板和从板)上，拥有64个VLSI处理器，每秒提供15亿个连接。它通过编程并行处理多层感知器型神经网络、反馈型神经网络(如Hopfield模型)和任何其他类型的神经网络的前向/后向计算。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A MULTINEURO computer system by using high performance digital neuro processor

We have developed an accelerator device "MULTINEURO-A" which incorporates multiple processors (one master node and plural slave nodes) dedicated to neural network high-speed computing realized on an engineering workstation. Nodes are connected by two data paths, broadcast bus and ring bus. Processing speed is in proportion to the number of processors. It is built in two boards (master and slave) having 64 VLSI processors that offers to 1.5 billion connections per seconds. It processes forward/backward calculation of multilayer perceptron type neural networks, feedback type neural networks such as Hopfield model, and any other types by programming for parallel processing.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers

自引率

0.00%

发文量