{"title":"一种针对具体应用的NBTI老化分析方法","authors":"H. Abbas, Mark Zwolinski, Basel Halak","doi":"10.1109/VARI.2015.7456553","DOIUrl":null,"url":null,"abstract":"There is growing concern about time-dependent performance variations of CMOS devices due to ageing-induced delay degradation. One of the main causes of ageing is Negative Bias Temperature Instability (NBTI). Existing models which predict the impact of NBTI on overall system performance assume a generic stress-recovery ratio of input signals of 50%. Such an assumption can cause misleading predictions about how a circuit's performance will degrade over time and more importantly which parts of the system will be most affected. This work develops a novel NBTI ageing analysis which is based on accurate calculations of the stress-recovery ratios for applicationspecific systems. The proposed method is employed to predict the ageing of an ARM processor synthesised to 90nm technology. Our results show the proposed ageing analysis techniques can significantly reduce prediction errors (e.g. 39% for one of the critical paths) compared to the generic models, it can also identify more accurately the parts of the system which are most vulnerable to ageing.","PeriodicalId":299950,"journal":{"name":"2015 International Workshop on CMOS Variability (VARI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"An application-specific NBTI ageing analysis method\",\"authors\":\"H. Abbas, Mark Zwolinski, Basel Halak\",\"doi\":\"10.1109/VARI.2015.7456553\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"There is growing concern about time-dependent performance variations of CMOS devices due to ageing-induced delay degradation. One of the main causes of ageing is Negative Bias Temperature Instability (NBTI). Existing models which predict the impact of NBTI on overall system performance assume a generic stress-recovery ratio of input signals of 50%. Such an assumption can cause misleading predictions about how a circuit's performance will degrade over time and more importantly which parts of the system will be most affected. This work develops a novel NBTI ageing analysis which is based on accurate calculations of the stress-recovery ratios for applicationspecific systems. The proposed method is employed to predict the ageing of an ARM processor synthesised to 90nm technology. Our results show the proposed ageing analysis techniques can significantly reduce prediction errors (e.g. 39% for one of the critical paths) compared to the generic models, it can also identify more accurately the parts of the system which are most vulnerable to ageing.\",\"PeriodicalId\":299950,\"journal\":{\"name\":\"2015 International Workshop on CMOS Variability (VARI)\",\"volume\":\"11 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 International Workshop on CMOS Variability (VARI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/VARI.2015.7456553\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 International Workshop on CMOS Variability (VARI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VARI.2015.7456553","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
An application-specific NBTI ageing analysis method
There is growing concern about time-dependent performance variations of CMOS devices due to ageing-induced delay degradation. One of the main causes of ageing is Negative Bias Temperature Instability (NBTI). Existing models which predict the impact of NBTI on overall system performance assume a generic stress-recovery ratio of input signals of 50%. Such an assumption can cause misleading predictions about how a circuit's performance will degrade over time and more importantly which parts of the system will be most affected. This work develops a novel NBTI ageing analysis which is based on accurate calculations of the stress-recovery ratios for applicationspecific systems. The proposed method is employed to predict the ageing of an ARM processor synthesised to 90nm technology. Our results show the proposed ageing analysis techniques can significantly reduce prediction errors (e.g. 39% for one of the critical paths) compared to the generic models, it can also identify more accurately the parts of the system which are most vulnerable to ageing.