Tomohiro Ueno, Takaaki Miyajima, Antoniette Mondigo, K. Sano
{"title":"Hybrid Network Utilization for Efficient Communication in a Tightly Coupled FPGA Cluster","authors":"Tomohiro Ueno, Takaaki Miyajima, Antoniette Mondigo, K. Sano","doi":"10.1109/ICFPT47387.2019.00068","DOIUrl":null,"url":null,"abstract":"A tightly coupled FPGA cluster is a promising approach for large-scale parallel processing with application specialized hardware. Along with the advantages of FPGA-based custom computing, such as high power efficiency, a customized network subsystem with efficient communication through direct Inter-FPGA links allows an FPGA cluster to be an effective platform for large-scale parallel processing. However, the cluster can suffer from substantial communication costs when a cluster becomes larger to obtain higher computing performance. In this paper, we propose to exploit the communication capacity of a host server network to improve communication performance. Besides, we show estimations for practical communication patterns on a network model in which we efficiently use both the FPGA and the host networks.","PeriodicalId":241340,"journal":{"name":"2019 International Conference on Field-Programmable Technology (ICFPT)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Field-Programmable Technology (ICFPT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICFPT47387.2019.00068","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
A tightly coupled FPGA cluster is a promising approach for large-scale parallel processing with application specialized hardware. Along with the advantages of FPGA-based custom computing, such as high power efficiency, a customized network subsystem with efficient communication through direct Inter-FPGA links allows an FPGA cluster to be an effective platform for large-scale parallel processing. However, the cluster can suffer from substantial communication costs when a cluster becomes larger to obtain higher computing performance. In this paper, we propose to exploit the communication capacity of a host server network to improve communication performance. Besides, we show estimations for practical communication patterns on a network model in which we efficiently use both the FPGA and the host networks.