{"title":"Application of data-flow computer to numerical simulation","authors":"S. Kishi","doi":"10.1109/IECON.1989.69728","DOIUrl":null,"url":null,"abstract":"Numerical calculations for the simulation of natural phenomena were conducted using a data-flow-type parallel processing computer. The aim of the study was to introduce an innovative software technology to the field of natural disaster prevention. In the study, two simulation models for natural phenomena were adopted as examples: one was a geological problem involving the solution of a partial differential equation by the difference calculus to study the behavior of a wetting front in the ground caused by rain-water in time series, while the other was a seismological problem of computing multiple integrals in the calculus of variations to seek the distribution of dislocations in a fault plane as an elastic body. These examples were programmed in Fortran in the usual computer and then, according to the sequential algorithm, translated into the data-flow assembler. The computing time in the data-flow computer was approximately three-to-five times shorter than that of the usual medium-size computer of computing speed 3 MIPS (million instructions per second). Dynamic visualization of the computing process was realized using an image display directly connected to the memory of the data-flow computer.<<ETX>>","PeriodicalId":384081,"journal":{"name":"15th Annual Conference of IEEE Industrial Electronics Society","volume":"667 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1989-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"15th Annual Conference of IEEE Industrial Electronics Society","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IECON.1989.69728","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Numerical calculations for the simulation of natural phenomena were conducted using a data-flow-type parallel processing computer. The aim of the study was to introduce an innovative software technology to the field of natural disaster prevention. In the study, two simulation models for natural phenomena were adopted as examples: one was a geological problem involving the solution of a partial differential equation by the difference calculus to study the behavior of a wetting front in the ground caused by rain-water in time series, while the other was a seismological problem of computing multiple integrals in the calculus of variations to seek the distribution of dislocations in a fault plane as an elastic body. These examples were programmed in Fortran in the usual computer and then, according to the sequential algorithm, translated into the data-flow assembler. The computing time in the data-flow computer was approximately three-to-five times shorter than that of the usual medium-size computer of computing speed 3 MIPS (million instructions per second). Dynamic visualization of the computing process was realized using an image display directly connected to the memory of the data-flow computer.<>