Impact of intrinsic parameter fluctuation on the fault tolerance of L1 data cache

R. A. Ahmed, K. Samsudin, F. Rokhani
{"title":"Impact of intrinsic parameter fluctuation on the fault tolerance of L1 data cache","authors":"R. A. Ahmed, K. Samsudin, F. Rokhani","doi":"10.1109/ICM.2009.5418676","DOIUrl":null,"url":null,"abstract":"As the semiconductor process technology continues to scale deeper into the nanometer region, the intrinsic parameter fluctuations will aggressively affect the performance and reliability of future microprocessors and System-on-Chip (SoC) applications. These system requires large SRAM arrays that occupy an increasing fraction of the chip real estate. To investigate the impact various source of intrinsic parameter fluctuation (IPF) from systems point of view, a framework to bridge architecture-level and device-level simulation will be utilized for data cache built from transistors with 25 nm, 18 nm and 13 nm technology node. This study found that the IPF will not have any significant impacts on data cache memory systems build with 25 nm while increasing the memory cell ratio, (β) to two will overcome the IPF impacts for the 18 nm. However, the 13 nm technology data cache could not operate even with higher cell ratio. Common, cache memory fault detection and correction such as ECC and redundancy can only partially remove the transaction error caused by these fluctuation sources.","PeriodicalId":391668,"journal":{"name":"2009 International Conference on Microelectronics - ICM","volume":"34 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 International Conference on Microelectronics - ICM","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICM.2009.5418676","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

Abstract

As the semiconductor process technology continues to scale deeper into the nanometer region, the intrinsic parameter fluctuations will aggressively affect the performance and reliability of future microprocessors and System-on-Chip (SoC) applications. These system requires large SRAM arrays that occupy an increasing fraction of the chip real estate. To investigate the impact various source of intrinsic parameter fluctuation (IPF) from systems point of view, a framework to bridge architecture-level and device-level simulation will be utilized for data cache built from transistors with 25 nm, 18 nm and 13 nm technology node. This study found that the IPF will not have any significant impacts on data cache memory systems build with 25 nm while increasing the memory cell ratio, (β) to two will overcome the IPF impacts for the 18 nm. However, the 13 nm technology data cache could not operate even with higher cell ratio. Common, cache memory fault detection and correction such as ECC and redundancy can only partially remove the transaction error caused by these fluctuation sources.
本征参数波动对L1数据缓存容错性的影响
随着半导体工艺技术不断深入纳米领域,其固有参数波动将严重影响未来微处理器和片上系统(SoC)应用的性能和可靠性。这些系统需要大型SRAM阵列,占据越来越多的芯片空间。为了从系统的角度研究不同来源的内在参数波动(IPF)对数据缓存的影响,我们将利用一个框架来桥接架构级和器件级的模拟,以25nm、18nm和13nm技术节点的晶体管构建数据缓存。本研究发现,IPF不会对25 nm的数据缓存系统产生任何显著影响,而增加存储单元比(β)到2将克服18 nm的IPF影响。然而,13纳米技术的数据缓存即使在更高的单元比下也无法运行。常见的cache故障检测和纠错,如ECC、冗余等,只能部分消除这些波动源造成的交易错误。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信