Sven Eisenhardt, Anja Küster, Thomas Schweizer, T. Kuhn, W. Rosenstiel
{"title":"用于容错的粗粒度可重构架构的运行时数据路径重新映射","authors":"Sven Eisenhardt, Anja Küster, Thomas Schweizer, T. Kuhn, W. Rosenstiel","doi":"10.1109/ReCoSoC.2011.5981536","DOIUrl":null,"url":null,"abstract":"In this contribution we present an approach which significantly enhances the reliability of coarse-grained reconfigurable architectures by applying a novel remapping strategy. If a component of the architecture is affected by a permanent failure, it will be deactivated and the architecture is reconfigured to relinquish the concerned resource. In our experiments, we have regarded the failure of each single component in the PE array. All of the failures were successfully repaired by applying our remapping method. In average, the repair of one failure took 6 minutes, and the clock frequency had to be reduced by just 0.6% to enable the execution of the changed application mapping.","PeriodicalId":103130,"journal":{"name":"6th International Workshop on Reconfigurable Communication-Centric Systems-on-Chip (ReCoSoC)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2011-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"Runtime-datapath-remapping for fault-tolerant coarse-grained reconfigurable architectures\",\"authors\":\"Sven Eisenhardt, Anja Küster, Thomas Schweizer, T. Kuhn, W. Rosenstiel\",\"doi\":\"10.1109/ReCoSoC.2011.5981536\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this contribution we present an approach which significantly enhances the reliability of coarse-grained reconfigurable architectures by applying a novel remapping strategy. If a component of the architecture is affected by a permanent failure, it will be deactivated and the architecture is reconfigured to relinquish the concerned resource. In our experiments, we have regarded the failure of each single component in the PE array. All of the failures were successfully repaired by applying our remapping method. In average, the repair of one failure took 6 minutes, and the clock frequency had to be reduced by just 0.6% to enable the execution of the changed application mapping.\",\"PeriodicalId\":103130,\"journal\":{\"name\":\"6th International Workshop on Reconfigurable Communication-Centric Systems-on-Chip (ReCoSoC)\",\"volume\":\"46 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2011-06-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"6th International Workshop on Reconfigurable Communication-Centric Systems-on-Chip (ReCoSoC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ReCoSoC.2011.5981536\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"6th International Workshop on Reconfigurable Communication-Centric Systems-on-Chip (ReCoSoC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ReCoSoC.2011.5981536","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Runtime-datapath-remapping for fault-tolerant coarse-grained reconfigurable architectures
In this contribution we present an approach which significantly enhances the reliability of coarse-grained reconfigurable architectures by applying a novel remapping strategy. If a component of the architecture is affected by a permanent failure, it will be deactivated and the architecture is reconfigured to relinquish the concerned resource. In our experiments, we have regarded the failure of each single component in the PE array. All of the failures were successfully repaired by applying our remapping method. In average, the repair of one failure took 6 minutes, and the clock frequency had to be reduced by just 0.6% to enable the execution of the changed application mapping.