D. Rodohan, S. Saunders, S. Cvetkovic, P. Beavis, R. Glover
{"title":"Parallel simulation of electromagnetic fields for telecommunication applications","authors":"D. Rodohan, S. Saunders, S. Cvetkovic, P. Beavis, R. Glover","doi":"10.1109/ICCS.1994.474215","DOIUrl":null,"url":null,"abstract":"The finite difference time domain (FDTD) algorithm has become a popular-technique with which to solve many different electromagnetic problems. However, for real problems the mesh size becomes large and as a result the computation time on a workstation is in the region of tens of hours. In fact, in some cases there will be insufficient memory available on a workstation to solve large problems. We describe a general method based on distributed computing to reduce both the memory requirements and the computation time of the FDTD algorithm. An implementation of the FDTD algorithm on a network of workstations is then evaluated. The implementation is shown to give a peak speed up of 7.5 on a network of 8 workstations over the sequential computation time.<<ETX>>","PeriodicalId":158681,"journal":{"name":"Proceedings of ICCS '94","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1994-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of ICCS '94","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCS.1994.474215","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 4
Abstract
The finite difference time domain (FDTD) algorithm has become a popular-technique with which to solve many different electromagnetic problems. However, for real problems the mesh size becomes large and as a result the computation time on a workstation is in the region of tens of hours. In fact, in some cases there will be insufficient memory available on a workstation to solve large problems. We describe a general method based on distributed computing to reduce both the memory requirements and the computation time of the FDTD algorithm. An implementation of the FDTD algorithm on a network of workstations is then evaluated. The implementation is shown to give a peak speed up of 7.5 on a network of 8 workstations over the sequential computation time.<>