{"title":"Reconfigurable multiple-behavioural architecture using free-space optical communication","authors":"Tut San Guan, S. Barros","doi":"10.1109/MPPOI.1994.336615","DOIUrl":null,"url":null,"abstract":"We describe a fully configurable, parallel architecture based on free-space optical communication. The noninterference and free-space transmission of light are exploited to provide parallel and arbitrary communications both within and among the processors. Matrix-addressed SLMs (spatial light modulators) are used to develop a computer model of the proposed scheme. Special instructions allowing users to configure both the topology and behaviour of the architecture are described. The resulting configurations faithfully reproduce the target architectures, including dataflow, pipelining, neural network, SIMD and MIMD. The use of optics also alleviates many communication bottlenecks usually associated with conventional computers. The architecture is proposed as the basis for a new kind of general/all purpose parallel computers.<<ETX>>","PeriodicalId":254893,"journal":{"name":"First International Workshop on Massively Parallel Processing Using Optical Interconnections","volume":"27 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1994-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"First International Workshop on Massively Parallel Processing Using Optical Interconnections","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MPPOI.1994.336615","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
We describe a fully configurable, parallel architecture based on free-space optical communication. The noninterference and free-space transmission of light are exploited to provide parallel and arbitrary communications both within and among the processors. Matrix-addressed SLMs (spatial light modulators) are used to develop a computer model of the proposed scheme. Special instructions allowing users to configure both the topology and behaviour of the architecture are described. The resulting configurations faithfully reproduce the target architectures, including dataflow, pipelining, neural network, SIMD and MIMD. The use of optics also alleviates many communication bottlenecks usually associated with conventional computers. The architecture is proposed as the basis for a new kind of general/all purpose parallel computers.<>