{"title":"SecCheck:一个具有不可信组件的可信系统","authors":"Rajshekar Kalayappan, S. Sarangi","doi":"10.1109/ISVLSI.2016.31","DOIUrl":null,"url":null,"abstract":"Mission critical applications face a security risk when they use third-party ICs for their speed and/or technology benefits. SecCheck is an architectural framework that securely incorporates fast, untrusted third-party cores (3PCs). It takes a comprehensive approach, providing for all of the different traditional fault tolerance techniques, to verify the 3PCs' functioning. The verification is done at run-time by slow, trusted, homegrown cores (HGCs). The overhead of providing security is reduced through intelligent scheduling exploiting task-level parallelism. The average performance penalty for achieving security under SecCheck is just 10-17% (optimal schedule), even when the HGCs are only half as fast as the 3PCs. We also devise a heuristic-based scheduler that is 500X faster than an ILP-based optimal one, with a relative penalty less than 1%.","PeriodicalId":140647,"journal":{"name":"2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"SecCheck: A Trustworthy System with Untrusted Components\",\"authors\":\"Rajshekar Kalayappan, S. Sarangi\",\"doi\":\"10.1109/ISVLSI.2016.31\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Mission critical applications face a security risk when they use third-party ICs for their speed and/or technology benefits. SecCheck is an architectural framework that securely incorporates fast, untrusted third-party cores (3PCs). It takes a comprehensive approach, providing for all of the different traditional fault tolerance techniques, to verify the 3PCs' functioning. The verification is done at run-time by slow, trusted, homegrown cores (HGCs). The overhead of providing security is reduced through intelligent scheduling exploiting task-level parallelism. The average performance penalty for achieving security under SecCheck is just 10-17% (optimal schedule), even when the HGCs are only half as fast as the 3PCs. We also devise a heuristic-based scheduler that is 500X faster than an ILP-based optimal one, with a relative penalty less than 1%.\",\"PeriodicalId\":140647,\"journal\":{\"name\":\"2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)\",\"volume\":\"49 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISVLSI.2016.31\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISVLSI.2016.31","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
SecCheck: A Trustworthy System with Untrusted Components
Mission critical applications face a security risk when they use third-party ICs for their speed and/or technology benefits. SecCheck is an architectural framework that securely incorporates fast, untrusted third-party cores (3PCs). It takes a comprehensive approach, providing for all of the different traditional fault tolerance techniques, to verify the 3PCs' functioning. The verification is done at run-time by slow, trusted, homegrown cores (HGCs). The overhead of providing security is reduced through intelligent scheduling exploiting task-level parallelism. The average performance penalty for achieving security under SecCheck is just 10-17% (optimal schedule), even when the HGCs are only half as fast as the 3PCs. We also devise a heuristic-based scheduler that is 500X faster than an ILP-based optimal one, with a relative penalty less than 1%.