Design of ultra high density and low power computational blocks using nano-magnets

M. Sharad, K. Yogendra, K. Kwon, K. Roy
{"title":"Design of ultra high density and low power computational blocks using nano-magnets","authors":"M. Sharad, K. Yogendra, K. Kwon, K. Roy","doi":"10.1109/ISQED.2013.6523614","DOIUrl":null,"url":null,"abstract":"All Spin Logic (ASL) employs multiple nano-magnets interacting through spin-torque using metallic interconnect. ASL gates, being magneto-metallic, can operate at ultra low terminal voltage of few millivolts, and hence can be exploited for low power computation. Since, nano-magnets can preserve their state upon withdrawal of supply voltage, ASL can be pipelined for higher performance, without insertion of extra latches. However, pipelining requires the use of clocked CMOS transistors, which significantly increase the required supply voltage. In this work we analyse the design of an 8-bit, pipelined ASL multiplier, integrated with CMOS clocking circuitry. We propose a design scheme for 3-D ASL, which involves stacking of multiple ASL layers that are clocked using the same CMOS transistors. Stacking of N ASL layers using the proposed scheme can enhance the power saving as well as area density by factor of N. The proposed design scheme for magneto-metallic computational blocks can achieve more than two order of magnitude higher density and 10x lower power consumption as compared to 15nm CMOS design.","PeriodicalId":127115,"journal":{"name":"International Symposium on Quality Electronic Design (ISQED)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2013-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"17","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Symposium on Quality Electronic Design (ISQED)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISQED.2013.6523614","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 17

Abstract

All Spin Logic (ASL) employs multiple nano-magnets interacting through spin-torque using metallic interconnect. ASL gates, being magneto-metallic, can operate at ultra low terminal voltage of few millivolts, and hence can be exploited for low power computation. Since, nano-magnets can preserve their state upon withdrawal of supply voltage, ASL can be pipelined for higher performance, without insertion of extra latches. However, pipelining requires the use of clocked CMOS transistors, which significantly increase the required supply voltage. In this work we analyse the design of an 8-bit, pipelined ASL multiplier, integrated with CMOS clocking circuitry. We propose a design scheme for 3-D ASL, which involves stacking of multiple ASL layers that are clocked using the same CMOS transistors. Stacking of N ASL layers using the proposed scheme can enhance the power saving as well as area density by factor of N. The proposed design scheme for magneto-metallic computational blocks can achieve more than two order of magnitude higher density and 10x lower power consumption as compared to 15nm CMOS design.
利用纳米磁体设计超高密度低功耗计算块
所有自旋逻辑(ASL)采用多个纳米磁体通过金属互连通过自旋扭矩相互作用。ASL门是磁性金属,可以在几毫伏的超低端电压下工作,因此可以用于低功耗计算。由于纳米磁铁可以在电源电压撤出时保持其状态,因此ASL可以流水线化以获得更高的性能,而无需插入额外的锁存器。然而,流水线需要使用带时钟的CMOS晶体管,这大大增加了所需的电源电压。在这项工作中,我们分析了一个集成了CMOS时钟电路的8位流水线ASL乘法器的设计。我们提出了一种三维ASL的设计方案,该方案涉及使用相同CMOS晶体管进行时钟的多个ASL层的堆叠。利用该方案叠加N个ASL层,可以将功耗和面积密度提高N倍。与15nm CMOS设计相比,该方案的磁金属计算块密度提高了两个数量级以上,功耗降低了10倍。
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:481959085
Book学术官方微信