An ultra high density pseudo dual-port SRAM in 16nm FINFET process for graphics processors

V. Nautiyal, G. Singla, Lalita Gupta, S. Dwivedi, M. Kinkade
{"title":"An ultra high density pseudo dual-port SRAM in 16nm FINFET process for graphics processors","authors":"V. Nautiyal, G. Singla, Lalita Gupta, S. Dwivedi, M. Kinkade","doi":"10.1109/SOCC.2017.8225996","DOIUrl":null,"url":null,"abstract":"In recent times, graphic, audio and video definition has improved due to significant advancement in complex algorithms and video processing techniques. These techniques require heterogeneous and multi-core processors because of their complex computation abilities. Dual-port memories have become an essential component of CPUs because multi-core processors require significant data transfer. However, dual-port memories come at a cost of increased area and leakage. In this paper, an ultra-high-density dual-port SRAM (RADPUHD) architecture is proposed which addresses area and leakage challenges. It is designed and fabricated in 16nm technology. This paper presents use of a single-port bitcell to achieve functionality of dual-port SRAM thus improving area efficiency. The use of latches for Port B signals instead of full flip-flops further reduces area. The proposed design is a bolt-on wrapper around a 6T single-port SRAM. This design achieved a memory density of 8.1Mb/mm2 chip area and achieved 53% area savings and approximately 60% leakage savings when compared to an 8T dual-port SRAM that was also fabricated in 16nm. Silicon results show that the proposed circuit is functional down to a minimum operating voltage of 520mV.","PeriodicalId":366264,"journal":{"name":"2017 30th IEEE International System-on-Chip Conference (SOCC)","volume":"546 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 30th IEEE International System-on-Chip Conference (SOCC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SOCC.2017.8225996","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7

Abstract

In recent times, graphic, audio and video definition has improved due to significant advancement in complex algorithms and video processing techniques. These techniques require heterogeneous and multi-core processors because of their complex computation abilities. Dual-port memories have become an essential component of CPUs because multi-core processors require significant data transfer. However, dual-port memories come at a cost of increased area and leakage. In this paper, an ultra-high-density dual-port SRAM (RADPUHD) architecture is proposed which addresses area and leakage challenges. It is designed and fabricated in 16nm technology. This paper presents use of a single-port bitcell to achieve functionality of dual-port SRAM thus improving area efficiency. The use of latches for Port B signals instead of full flip-flops further reduces area. The proposed design is a bolt-on wrapper around a 6T single-port SRAM. This design achieved a memory density of 8.1Mb/mm2 chip area and achieved 53% area savings and approximately 60% leakage savings when compared to an 8T dual-port SRAM that was also fabricated in 16nm. Silicon results show that the proposed circuit is functional down to a minimum operating voltage of 520mV.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
一种用于图形处理器的16nm FINFET工艺的超高密度伪双端口SRAM
近年来,由于复杂算法和视频处理技术的显著进步,图形、音频和视频清晰度得到了改善。这些技术需要异构和多核处理器,因为它们具有复杂的计算能力。由于多核处理器需要大量的数据传输,双端口存储器已成为cpu的重要组成部分。然而,双端口存储器的代价是增加面积和泄漏。本文提出了一种解决面积和泄漏问题的超高密度双端口SRAM (RADPUHD)架构。它采用16nm技术设计和制造。本文提出使用单端口位元来实现双端口SRAM的功能,从而提高面积效率。使用锁存器的端口B信号,而不是完整的触发器进一步减少面积。提出的设计是围绕6T单端口SRAM的螺栓连接包装。该设计实现了8.1Mb/mm2芯片面积的内存密度,与同样采用16nm工艺制造的8T双端口SRAM相比,节省了53%的面积和大约60%的泄漏。硅测试结果表明,该电路的工作电压可低至520mV。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Content-aware line-based power modeling methodology for image signal processor Power and area evaluation of a fault-tolerant network-on-chip A low-pass continuous-time delta-sigma interface circuit for wideband MEMS gyroscope readout ASIC Lithography hotspot detection: From shallow to deep learning The path to global connectivity — Wireless communication enters the next generation
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1