{"title":"基于sram的fpga中容错无晶圆clb的比较","authors":"Arwa Ben Dhia, L. Naviner, Philippe Matherat","doi":"10.1109/LATW.2013.6562661","DOIUrl":null,"url":null,"abstract":"This paper first introduces three new architectures of a voter for a Butterfly CLB in an SRAM-based FPGA. Taking into account area and delay constraints, an optimized voter architecture is picked up for the Butterfly design. Another version for the latter is also proposed in order to reduce the area overhead. Afterward, we synthesize different CLB fabless architectures in STM 65nm CMOS technology and evaluate their fault tolerance and reliability, so as to obtain an overview of the current state of the art. Finally, we compare the CLB architectures with respect to the conventional one by defining a metric expressing the tradeoff between the fault tolerance gain, the performance degradation and the cost penalties including area, power and SRAM memory.","PeriodicalId":186736,"journal":{"name":"2013 14th Latin American Test Workshop - LATW","volume":"85 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Comparison of fault-tolerant fabless CLBs in SRAM-based FPGAs\",\"authors\":\"Arwa Ben Dhia, L. Naviner, Philippe Matherat\",\"doi\":\"10.1109/LATW.2013.6562661\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper first introduces three new architectures of a voter for a Butterfly CLB in an SRAM-based FPGA. Taking into account area and delay constraints, an optimized voter architecture is picked up for the Butterfly design. Another version for the latter is also proposed in order to reduce the area overhead. Afterward, we synthesize different CLB fabless architectures in STM 65nm CMOS technology and evaluate their fault tolerance and reliability, so as to obtain an overview of the current state of the art. Finally, we compare the CLB architectures with respect to the conventional one by defining a metric expressing the tradeoff between the fault tolerance gain, the performance degradation and the cost penalties including area, power and SRAM memory.\",\"PeriodicalId\":186736,\"journal\":{\"name\":\"2013 14th Latin American Test Workshop - LATW\",\"volume\":\"85 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-04-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2013 14th Latin American Test Workshop - LATW\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/LATW.2013.6562661\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 14th Latin American Test Workshop - LATW","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/LATW.2013.6562661","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Comparison of fault-tolerant fabless CLBs in SRAM-based FPGAs
This paper first introduces three new architectures of a voter for a Butterfly CLB in an SRAM-based FPGA. Taking into account area and delay constraints, an optimized voter architecture is picked up for the Butterfly design. Another version for the latter is also proposed in order to reduce the area overhead. Afterward, we synthesize different CLB fabless architectures in STM 65nm CMOS technology and evaluate their fault tolerance and reliability, so as to obtain an overview of the current state of the art. Finally, we compare the CLB architectures with respect to the conventional one by defining a metric expressing the tradeoff between the fault tolerance gain, the performance degradation and the cost penalties including area, power and SRAM memory.