{"title":"物理设计空间探索","authors":"Ephrem Wu, Inkeun Cho","doi":"10.1145/2684746.2689080","DOIUrl":null,"url":null,"abstract":"A polynomial accelerator implemented with a custom high-dynamic-range number representation operates up to 534MHz in the slowest speed grade on a 28nm FPGA, a clock rate that a typical FPGA tool flow cannot achieve. This design tutorial shows how to achieve a physically scalable and high-speed numerical design by partitioning it into a cascade of identical stages, and balancing the LUT-to-DSP ratio within each stage to match the available resources on the FPGA.","PeriodicalId":388546,"journal":{"name":"Proceedings of the 2015 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","volume":"23 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Physical Design Space Exploration\",\"authors\":\"Ephrem Wu, Inkeun Cho\",\"doi\":\"10.1145/2684746.2689080\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A polynomial accelerator implemented with a custom high-dynamic-range number representation operates up to 534MHz in the slowest speed grade on a 28nm FPGA, a clock rate that a typical FPGA tool flow cannot achieve. This design tutorial shows how to achieve a physically scalable and high-speed numerical design by partitioning it into a cascade of identical stages, and balancing the LUT-to-DSP ratio within each stage to match the available resources on the FPGA.\",\"PeriodicalId\":388546,\"journal\":{\"name\":\"Proceedings of the 2015 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays\",\"volume\":\"23 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-02-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 2015 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2684746.2689080\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2015 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2684746.2689080","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A polynomial accelerator implemented with a custom high-dynamic-range number representation operates up to 534MHz in the slowest speed grade on a 28nm FPGA, a clock rate that a typical FPGA tool flow cannot achieve. This design tutorial shows how to achieve a physically scalable and high-speed numerical design by partitioning it into a cascade of identical stages, and balancing the LUT-to-DSP ratio within each stage to match the available resources on the FPGA.